WO2025259282A1 - Content for displaying in an ambient mode screen for a television application - Google Patents

Content for displaying in an ambient mode screen for a television application

Info

Publication number
WO2025259282A1
WO2025259282A1 PCT/US2024/033840 US2024033840W WO2025259282A1 WO 2025259282 A1 WO2025259282 A1 WO 2025259282A1 US 2024033840 W US2024033840 W US 2024033840W WO 2025259282 A1 WO2025259282 A1 WO 2025259282A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
keyword
display device
data
content information
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
PCT/US2024/033840
Other languages
French (fr)
Inventor
Sayantan Paul
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Google LLC
Original Assignee
Google LLC
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Google LLC filed Critical Google LLC
Priority to PCT/US2024/033840 priority Critical patent/WO2025259282A1/en
Publication of WO2025259282A1 publication Critical patent/WO2025259282A1/en
Pending legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • H04N21/431—Generation of visual interfaces for content selection or interaction; Content or additional data rendering
    • H04N21/4312—Generation of visual interfaces for content selection or interaction; Content or additional data rendering involving specific graphical features, e.g. screen layout, special fonts or colors, blinking icons, highlights or animations
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/047—Probabilistic or stochastic networks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/00—Two-dimensional [2D] image generation
    • G06T11/60—Creating or editing images; Combining images with text
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
    • H04N21/235—Processing of additional data, e.g. scrambling of additional data or processing content descriptors
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
    • H04N21/251—Learning process for intelligent management, e.g. learning user preferences for recommending movies
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
    • H04N21/258—Client or end-user data management, e.g. managing client capabilities, user preferences or demographics, processing of multiple end-users preferences to derive collaborative data
    • H04N21/25866—Management of end-user data
    • H04N21/25891—Management of end-user data being end-user preferences
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • H04N21/435—Processing of additional data, e.g. decrypting of additional data, reconstructing software from modules extracted from the transport stream
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
    • H04N21/81—Monomedia components thereof
    • H04N21/8126—Monomedia components thereof involving additional data, e.g. news, sports, stocks, weather forecasts

Definitions

  • a television (TV) application may present various types of media content of interest to a user on a smart TV.
  • the TV application may display an ambient screen on the smart TV.
  • the ambient screen may be a blank screen.
  • the ambient screen may display an image that will not cause any additional wear, burn-in, or damage to the display of the smart TV.
  • a smart television in a room of a home of a user may be the centerpiece of the room due to its large size and prominent placement in the room.
  • the smart TV may provide an ambient screen or a home screen when the user is not interacting with or watching media content on the smart TV. For example, when the user turns on the smart TV, the smart TV may provide the ambient screen until the user selects or clicks a button on a remote control for the smart TV.
  • the ambient screen may be a blank screen (no image is displayed on the screen of the smart TV).
  • a TV application running on the smart TV may display images in an iterative manner from a photo collection of the user.
  • the smart TV may be prominently placed in a room, the user may want to have the smart TV display an image on the ambient screen. Some users may want to have the smart TV display an interesting, artistic looking image on the ambient screen making the smart TV appear to be a piece of artwork in the room.
  • the TV application may provide additional information on the ambient screen that may include, but is not limited to, a time and date, current weather conditions, a news headline, a calendar reminder, and one or more minimalistic advertisements. Each information entry on the ambient screen may be selected or clicked on by the user interacting with a remote control for the smart TV. Once selected, each entry may direct the user to a web site or application on the smart TV.
  • selecting the news headline may direct the user to an application on the smart TV that provides news-based streaming media Atty Docket No.0120-996WO1 content.
  • selecting a minimalistic advertisement may direct the user to a web site to obtain more information related to the content of the advertisement (e.g., how to purchase an item).
  • the techniques described herein relate to a method including: sending, by a server computer to a display device, images for including in a carousel on the display device, the images included in the carousel for use as a basis for generating customized artwork for display in an ambient screen of a television application executing on the display device; generating an output image including: receiving a selection of an input image from the carousel; extracting image content information and data from the input image; determining keyword tokens for associating with the input image; and applying an ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens; and sending, by the server computer to the display device, the output image as the customized artwork for display in the ambient screen of the television application executing on the display device.
  • the techniques described herein relate to a method, wherein extracting image content information and data from the input image includes extracting the image content information and data from the input image using generative artificial intelligence.
  • the techniques described herein relate to a method, wherein generating the output image further includes applying a generative artificial intelligence model that includes an image comprehension model and an image generation model.
  • the techniques described herein relate to a method, wherein the ensemble of style transfer networks and generative adversarial networks are included in the image generation model.
  • the techniques described herein relate to a method, wherein extracting image content information and data from the input image includes extracting the image content information and data from the input image by the image comprehension model.
  • the techniques described herein relate to a method, wherein the keyword tokens include an image comprehension keyword token; and wherein the image Atty Docket No.0120-996WO1 comprehension model determines the image comprehension keyword token for the input image based on the image content information and data for the input image. [0010] In some aspects, the techniques described herein relate to a method, wherein the image generation model generates the output image as an artistic rendering of the input image. [0011] In some aspects, the techniques described herein relate to a method, wherein the keyword tokens include an image comprehension keyword token and at least one additional keyword token; and wherein generating the output image further includes generating a homologated keyword token based on the image comprehension keyword token and the at least one additional keyword token.
  • the techniques described herein relate to a method, wherein applying the ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens includes applying the ensemble of style transfer networks and generative adversarial networks on the homologated keyword token. [0013] In some aspects, the techniques described herein relate to a method, wherein the output image is generated by an image-to-image synthesizer.
  • the techniques described herein relate to a non-transitory computer-readable medium storing executable instructions that when executed by at least one processor of a server computer cause the at least one processor to execute operations, the operations including: sending, to a display device, images for including in a carousel on the display device, the images included in the carousel for use as a basis for generating customized artwork for display in an ambient screen of a television application executing on the display device; generating an output image including: receiving a selection of an input image from the carousel; extracting image content information and data from the input image; determining keyword tokens for associating with the input image; and applying an ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens; and sending, to the display device, the output image as the customized artwork for display in the ambient screen of the television application executing on the display device.
  • the techniques described herein relate to a non-transitory computer-readable medium, wherein extracting image content information and data from the input image includes extracting the image content information and data from the input image using generative artificial intelligence.
  • the techniques described herein relate to a non-transitory computer-readable medium, wherein generating the output image further includes applying a generative artificial intelligence model that includes an image comprehension model and an image generation model.
  • the techniques described herein relate to a non-transitory computer-readable medium, wherein the ensemble of style transfer networks and generative adversarial networks are included in the image generation model.
  • the techniques described herein relate to a non-transitory computer-readable medium, wherein extracting image content information and data from the input image includes extracting the image content information and data from the input image by the image comprehension model.
  • the techniques described herein relate to a non-transitory computer-readable medium, wherein the keyword tokens include an image comprehension keyword token; and wherein the image comprehension model determines the image comprehension keyword token for the input image based on the image content information and data for the input image.
  • the techniques described herein relate to a non-transitory computer-readable medium, wherein the image generation model generates the output image as an artistic rendering of the input image.
  • the techniques described herein relate to a non-transitory computer-readable medium, wherein the keyword tokens include an image comprehension keyword token and at least one additional keyword token; and wherein generating the output image further includes generating a homologated keyword token based on the image comprehension keyword token and the at least one additional keyword token.
  • the techniques described herein relate to a non-transitory computer-readable medium, wherein applying the ensemble of style transfer networks and generative adversarial networks on the image content information and data using the Atty Docket No.0120-996WO1 keyword tokens includes applying the ensemble of style transfer networks and generative adversarial networks on the homologated keyword token.
  • the techniques described herein relate to a non-transitory computer-readable medium, wherein the output image is generated by an image-to-image synthesizer.
  • the techniques described herein relate to a system including: at least one processor; and a non-transitory computer-readable medium storing instructions that when executed by the at least one processor cause the system to: send, from a server computer to a display device, images for including in a carousel on the display device, the images included in the carousel for use as a basis for generating customized artwork for display in an ambient screen of a television application executing on the display device; generate an output image including: receiving a selection of an input image from the carousel; extracting image content information and data from the input image; determining keyword tokens for associating with the input image; and applying an ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens; and send, from the server computer to the display device, the output
  • the techniques described herein relate to a system, wherein extracting image content information and data from the input image includes extracting the image content information and data from the input image using generative artificial intelligence. [0026] In some aspects, the techniques described herein relate to a system, wherein generating the output image further includes applying a generative artificial intelligence model that includes an image comprehension model and an image generation model. [0027] In some aspects, the techniques described herein relate to a system, wherein the ensemble of style transfer networks and generative adversarial networks are included in the image generation model.
  • the techniques described herein relate to a system, wherein extracting image content information and data from the input image includes extracting the Atty Docket No.0120-996WO1 image content information and data from the input image by the image comprehension model.
  • the techniques described herein relate to a system, wherein the keyword tokens include an image comprehension keyword token; wherein the image comprehension model determines the image comprehension keyword token for the input image based on the image content information and data for the input image.
  • the techniques described herein relate to a system, wherein the image generation model generates the output image as an artistic rendering of the input image.
  • the techniques described herein relate to a system, wherein the keyword tokens include an image comprehension keyword token and at least one additional keyword token; and wherein generating the output image further includes generating a homologated keyword token based on the image comprehension keyword token and the at least one additional keyword token.
  • the techniques described herein relate to a system, wherein applying the ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens includes applying the ensemble of style transfer networks and generative adversarial networks on the homologated keyword token.
  • the techniques described herein relate to a system, wherein the output image is generated by an image-to-image synthesizer.
  • FIG.1A illustrates an example of a user interacting with a network-connected display device and a media adapter, according to implementations described throughout this disclosure.
  • FIG. 1B illustrates an example system for selecting, generating, and displaying an artistic-looking image on an ambient screen of a smart TV according to implementations described throughout this disclosure.
  • Atty Docket No.0120-996WO1 [0037] FIG.
  • FIG. 1C illustrates an example illustration of a process flow for selecting, generating, and displaying a customized artwork on an ambient screen of a smart TV according to implementations described throughout this disclosure.
  • FIG. 2 is an illustration of an example process for generating a customized artwork based on an input image according to implementations described throughout this disclosure.
  • FIG.3A is an illustration of a first image that includes three snowmen.
  • FIG.3B is an illustration of a second image that is based on and is an artistic rendering of a first image.
  • FIG.3C is an illustration of an ambient screen of a smart TV displaying an output image 204 as a customized artwork.
  • FIG. 4 is an illustration of another example process for generating and displaying a customized artwork on an ambient screen of a smart TV according to implementations described throughout this disclosure.
  • FIG.5 illustrates a flowchart depicting example operations of generating an image (e.g., a customized artwork) for display in an ambient screen of a television application executing on a display device according to implementations described throughout this disclosure.
  • DETAILED DESCRIPTION A user may place a smart television (TV) in a prominent spot in a room making the smart TV a focal point of the room. The user may want the smart TV, when not streaming media content for viewing by the user, to display an image on the ambient or home screen of the smart TV.
  • TV smart television
  • the user may want the smart TV, when not streaming media content for viewing by the user, to display an image on the ambient or home screen of the smart TV.
  • a television application running on a smart television may facilitate the displaying of the ambient screen image.
  • the ambient screen image may be a user selected image that may also include one or more entries associated with other information that may be of interest to the user.
  • the additional entries may be informative, selectable entries such as a current time and date, current weather conditions, a current news headline, a current calendar entry, and one or more minimalistic advertisements.
  • the user may want the image to be a pleasant, artistic looking image.
  • the smart TV may provide the ambient screen when a user is not interacting with or watching media content on the smart TV, for example, when the smart TV is first Atty Docket No.0120-996WO1 turned on.
  • the smart TV may provide and, in addition, update the content of the ambient screen until the user selects or clicks a button on a remote control for the smart TV that initiates another action on the smart TV. For example, selecting or clicking a guide button on the remote control may exit the ambient screen mode and the TV application may then display a channel guide to the user.
  • the user may use the remote control to select one of the entries on the ambient screen that then directs the user to web sites or applications on the smart TV. For example, selecting a news headline may direct the user to an application on the smart TV that provides streaming news media content. For example, selecting a minimalistic advertisement may direct the user to a web site to obtain more information for a product or service covered by the advertisement.
  • a server-side TV application may facilitate the selection, generation, and displaying of an image on the ambient screen of the smart TV.
  • the image may be an artistic-looking image.
  • an image-to-image synthesizer may utilize an ensemble based generative AI model that uses style transfer generative adversarial networks (GANS) along with generative artificial intelligence (AI) to create or generate a customized artwork based on particular image content for an image.
  • the image-to-image synthesizer may utilize Generative Adversarial Networks (GANs) when generating the customized artwork.
  • GANs are a type of machine learning algorithm that generates new, previously unseen data that is similar to existing data.
  • the fields of art and synthetic media may use GANs to create new and unique works of art and synthetic media that may be based on existing artworks.
  • style transfer is a technique common in the technical areas of computer vision and graphics that involves generating a new image by combining the content of one image with the style of another image.
  • Style transfer can be used when generating a custom artwork from an image by combining the image with the style of a selected work of art.
  • the use of such an ensemble based generative AI model can allow for the generating of an image more quickly, efficiently, and with fewer resources than previous approaches.
  • the customized artwork may be based on an image included in an image repository.
  • the image repository may include photos, stock images, and other types of images selected by or associated with the user.
  • the server-side TV application may provide or send images used as a basis for the customized artworks to a Atty Docket No.0120-996WO1 carousel of featured images included in the smart TV.
  • a television (TV) application executing on the smart TV may be configured to display a plurality of customized artworks generated from images included in the features carousel of images on an ambient or home screen of the TV application.
  • the TV application may display the customized artworks sequentially according to a temporal order of the images in the featured carousel of images.
  • the TV application may temporally organize the customized artworks as a carousel, smoothly transitioning from displaying one customized artwork after another customized artwork.
  • a mixer may automatically mix the images included in the carousel of featured images sequentially according to the temporal order.
  • an image-to-image synthesizer executing on the smart TV may generate the customized artwork on the smart TV utilizing the computing resources of the smart TV.
  • the server-side TV application may interface with the image-to-image synthesizer on the smart TV to provide or send image from an image repository to the image-to-image synthesizer for use in generating a customized artwork for the ambient screen of the smart TV.
  • At least one technical problem involves how to generate an image for display on an ambient or home screen of a smart TV.
  • At least one technical solution to the technical problem generates an image for display based on images included in an image repository for a user.
  • the image is generated by the technical process of applying an ensemble of style transfer networks and generative adversarial networks on the image content information and data using keyword tokens.
  • One implementation of the technical solution described herein uses generative artificial intelligence to extract image content from the featured carousel of images and an image-to-image synthesizer that utilizes an ensemble based generative AI model that uses style transfer generative adversarial networks (GANS) to create or generate a customized artwork.
  • GANS style transfer generative adversarial networks
  • At least one technical effect may be providing an alternative way to generate an image for display on an ambient or home screen of a smart TV.
  • FIG. 1A illustrates an example of a user 101 interacting with a network- connected display device 104 and a media adapter 107 according to implementations described throughout this disclosure.
  • FIG. 1A illustrates an example of a user 101 interacting with a network- connected display device 104 and a media adapter 107 according to implementations described throughout this disclosure.
  • FIG. 1B illustrates an example system 100 for selecting, generating, and displaying an artistic-looking image (e.g., customized artwork 113) on an ambient screen (e.g., home or ambient screen 109) of a smart TV (e.g., the network-connected display device 104) according to implementations described throughout this disclosure.
  • FIG.1C illustrates an example illustration of a process flow 190 for selecting, generating, and displaying a customized artwork (e.g., customized artwork 113) on an ambient screen (e.g., home or ambient screen 109) of a smart TV (e.g., the network- connected display device 104) according to implementations described throughout this disclosure.
  • FIGS. 1B illustrates an example system 100 for selecting, generating, and displaying an artistic-looking image (e.g., customized artwork 113) on an ambient screen (e.g., home or ambient screen 109) of a smart TV (e.g., the network-connected display device 104) according to implementations described throughout this disclosure.
  • the user 101 may enter a room 115 that includes the network-connected display device 104.
  • the user 101 interacting with a remote control device 105 for the network-connected display device 104 may turn the network-connected display device 104 on.
  • the network-connected display device 104 may display the ambient screen 109.
  • the ambient screen 109 may include the customized artwork 113 and entries 111a-d.
  • Entry 111a shows the current date and time.
  • the user 101 may select or click on the entry 111a.
  • the network-connected display device 104 may run a date and time application on the network-connected display device 104.
  • Entry 111b shows the current temperature and weather for a location of the user 101.
  • the user 101 may select or click on the entry 111b.
  • the network- connected display device 104 may run a weather application on the network-connected display device 104.
  • Entry 111c shows a recent news headline.
  • the user 101 may select or click on the entry 111c.
  • the network-connected display device 104 may run a streaming news media application on the network-connected display device 104.
  • Entry 111d shows a minimalistic advertisement for a product. For example, the user 101 may select or click on the entry 111d.
  • the network-connected display device 104 may run a web browser application that may display a web site for the product as advertised in the entry 111d.
  • the date and time application, the weather application, the streaming news media application, and the web browser application may be included in television (TV) application(s) 136 on the network-connected display device 104.
  • the application(s) 136 may include media content provider applications, web browser applications, and other applications that may be installed on the network-connected display device 104 and accessed by the user 101.
  • the network-connected display device 104 may access one or more applications by way of a network 150.
  • the network-connected display device 104 may provide the user 101 with information and data provided by the applications to the user 101, for example, in a user interface 112 on the display 132 of the network-connected display device 104.
  • the network-connected display device 104 may communicate with a server computer 106 and media content providers 160 by way of the network 150.
  • the media content providers 160, the network-connected display device 104, the server computer 106, and a mobile computing device 102 may interact with and communicate with one other by way of the network 150.
  • the mobile computing device 102 may interface or connect to the media adapter 107 and/or the network-connected display device 104 by way of a wireless communication link that may be a short-range wireless connection such as, for example a Bluetooth connection or a Wi- Fi (e.g., direct Wi-Fi) connection.
  • a wireless communication link that may be a short-range wireless connection such as, for example a Bluetooth connection or a Wi- Fi (e.g., direct Wi-Fi) connection.
  • the user 101 may connect to and interact with a media adapter (e.g., the media adapter 107) by way of a network- connected display device (e.g., the network-connected display device 104) using a server- side television (TV) application (e.g., the server-side TV application 116) installed on a server computer (e.g., the server computer 106).
  • TV server- side television
  • the media adapter 107 may be connected or interfaced to the network-connected display device 104.
  • the network-connected display device 104 may be communicatively coupled or connected to the server computer 106 by way of the network 150.
  • a unified media platform (UMP) 158 may provide or serve media content items from the media content providers 160 to the network-connected display device 104 by way of the media adapter 107.
  • the network-connected display device 104 may Atty Docket No.0120-996WO1 execute a unified television application 130 that may interface with the server-side TV application 116 by way of the media adapter 107.
  • the server-side TV application 116 may interface with the unified television application 130 to facilitate providing the customized artwork 113 for the ambient screen 109 of the network-connected display device 104 when the network-connected display device 104 is turned on by the user 101.
  • the user 101 may interact with a network-connected display device (e.g., the network-connected display device 104) using a remote control device (e.g., the remote control device 105).
  • a television (TV) application 110 may render a virtual remote control 138 in a user interface (e.g., UI 114) on a display (e.g., a mobile computing device display 108) on the mobile computing device 102.
  • the virtual remote control 138 may allow the mobile computing device 102 to act as a remote control for the network-connected display device 104.
  • the TV application 110 may render the virtual remote control 138 for use with the network-connected display device 104.
  • the user 101 may interact with the remote control device 105 and/or the virtual remote control 138 when selecting media content for viewing on the network- connected display device 104.
  • the user 101 may interact with remote control device 105 and/or the virtual remote control 138 to select or click on and launch applications included in the TV application(s) 136.
  • the user 101 may interact with the remote control device 105 and/or the virtual remote control 138 to select or click on the customized artwork 113 displayed on the ambient screen 109 to lead the user 101 into a featured images carousel 134 that includes the content or image used as the basis for the customized artwork 113.
  • the user 101 may connect to and interact with a media adapter (e.g., the media adapter 107) using a TV application (e.g., the television (TV) application 110) installed on a mobile computing device (e.g., the mobile computing device 102).
  • a media adapter e.g., the media adapter 107
  • TV application e.g., the television (TV) application 110
  • a mobile computing device e.g., the mobile computing device 102
  • a user may connect to and interact with a media adapter (e.g., the media adapter 107) using a media adapter remote control device (e.g., media adapter remote control device 103).
  • the TV application 110 may render the virtual remote control 138 for use with the media adapter 107.
  • the virtual remote control 138 may allow the mobile computing Atty Docket No.0120-996WO1 device 102 to act as a remote control for the media adapter 107.
  • the user e.g., the user 101
  • the network-connected display device 104 may execute the unified television application 130.
  • the unified television application 130 may interface with a server-side television (TV) application 116.
  • the UMP 158 may interface with the unified television application 130 executing on the network-connected display device 104 and the media content providers 160 to provide media content to the network-connected display device 104.
  • the server computer 106 may include a knowledge module 166.
  • the knowledge module 166 may include information associated with media content items provided by the media content providers 160.
  • the knowledge module 166 may generate media content recommendations for associating with an account of a user based, in part, on a multi-dimensional user activity characteristic associated with the account of the user and the information associated with media content items provided by the media content providers 160.
  • the user activity characteristic associated with the account of the user may be obtained from a plurality of information sources that may include, but are not limited to, a search engine, a mapping application, and an online retailer.
  • the information sources may provide activity data related to activities of the account of the user by way of a respective software program or application.
  • the unified television application 130 may interface with the knowledge module 166 to provide information and data related to the past activities of the user when interacting with the unified television application 130, the viewing history of the user, and/or the popularity of media content items of a type, classification, category, group or genre.
  • the knowledge module 166 may help the unified television application 130 identify media content that may be useful and of interest to the user.
  • the unified television application 130 interfacing with the knowledge module 166 may curate or provide media content recommendations based on the past activities of the user when interacting with the unified television application 130, the viewing history of the user, and/or the popularity of media content items of a certain type, classification, category, group or genre.
  • the unified television application 130 may provide image(s) related to or associated Atty Docket No.0120-996WO1 with the recommended media content to an images repository 168.
  • the images repository 168 may include images associated with a user (e.g., the user 101) such as photographs and image files.
  • the server-side TV application 116 may facilitate providing or sending images from the images repository 168 to the featured images carousel 134 for use as home screens for a smart TV (e.g., the network-connected display device 104).
  • the image-to-image synthesizer 176 may receive images from the images repository 168.
  • the image-to-image synthesizer 176 may use an ensemble based generative AI model 178 included in generative artificial intelligence (Gen AI) model(s) 164 to extract information and data associated with the image (image content) from the images as a basis for generating artworks.
  • Gene AI generative artificial intelligence
  • the ensemble based generative AI model 178 may utilize generative artificial intelligence and style transfer based Generative Adversarial Networks (GANs) to extract content from images, to generate new data based on and that is similar to existing data for an artwork, and to create or generate a new image (e.g., a customized artwork) by combining the extracted image content of an image of the user (e.g., an image from the images repository 168) with the new data.
  • GANs Generative Adversarial Networks
  • the image-to-image synthesizer 176 may use GANs to create new and unique works of art and synthetic media that may be based on existing artworks. Style transfer techniques may generate a new image by combining the content of one image with the style of another image.
  • Combining the use of GANS with style transfer techniques may generate or create customized artworks for a user by combining an image selected by a user with the style of a work of art that is also selected by the user.
  • a technical benefit may be generating or creating the customized artworks on the smart TV itself by implementing the image-to-image synthesizer on the smart TV (e.g., the network-connected display device 104). This may result in a quicker more efficient way of generating or creating the customized artworks.
  • the unified television application 130 may present a user interface in the user interface 112 that allows the user (e.g., the user 101) to select one or more images from the images repository 168 that the user would like to use in a carousel of featured images for use as a basis for generating customized artworks for presenting in the ambient screen 109 of the network-connected display device 104.
  • the selected image may be provided as an input image.
  • the input image may be sent from the display device to the server computer.
  • Atty Docket No.0120-996WO1 [0063] In some implementations, referring to FIGS.
  • a customized artwork (e.g., the customized artwork 113) may be based on an image included in the images repository 168 and sent or provided to the featured images carousel 134.
  • the image repository may include photos, stock images, and other types of images selected by or associated with the user.
  • the server-side TV application 116 may provide or send images used as a basis for the customized artworks to the featured images carousel 134 included in the network-connected display device 104.
  • the unified television application 130 executing on the network-connected display device 104 may be configured to display multiple customized artworks based on images included in the featured images carousel 134 on the ambient screen 109.
  • the unified television application 130 may cause the generation and display of the customized artworks sequentially according to a temporal order of the images included in the featured images carousel 134.
  • the unified television application 130 may temporally organize the images and as such the customized artworks as a carousel, smoothly transitioning from the displaying of one customized artwork to another customized artwork.
  • the unified television application 130 may automatically mix the images included in the featured images carousel 134 sequentially according to the temporal order. Transitioning between customized artwork in a round robin fashion by accessing a carousel of featured images may reduce pixel burn-in and damage or wear to the display.
  • the unified television application 130 may present a user interface in the user interface 112 that allows a user (e.g., the user 101) to specify an artwork as the basis for the image-to-image synthesizing.
  • a user e.g., the user 101
  • the user may select an artwork by choosing from names of artworks by entering the name of an artwork in a text entry field.
  • the user may select an artwork for a list of artworks presented in a dropdown menu.
  • the names of artworks included for selection in the dropdown menu may be based on information about preferences of the user as provided by the knowledge module 166 (e.g., the user likes contemporary home furnishings and music so the artworks may be artworks by contemporary artists).
  • the user interface 112 may allow the user to enter the name of an artist (e.g., Degas) and then the user interface may present the user with a selection of artworks by the artist in a dropdown menu for the user to choose from.
  • the user may specify a different artwork for use in Atty Docket No.0120-996WO1 generating a customized artwork for each selected image from the images repository 168.
  • the user may specify a single artwork for use in creating a customized artwork for all of the user selected images from the images repository 168.
  • the server-side TV application 116 and the image-to-image synthesizer 176 may interface with the artificial intelligence (AI) module 194 that includes the generative artificial intelligence (Gen AI) model(s) 164 and a generative artificial intelligence (Gen AI) engine 162.
  • the Gen AI model(s) 164 may be machine learning trained models (e.g., the ensemble based generative AI model 178 described herein) for use by the Gen AI engine 162.
  • the Gen AI engine 162 can use generative artificial intelligence for extracting content from images included in the images repository 168.
  • the image-to-image synthesizer 176 may use the extracted content as a basis for generating or creating customized artwork for displaying as ambient screens (e.g., the ambient screen 109) on the network-connected display device 104 (e.g., a smart TV).
  • the image-to-image synthesizer 176 may interface with the Gen AI engine 162 to use the ensemble based generative AI model 178 on the extracted content from the images by applying the style transfer based GANs to the images to generate customized artworks based on the images.
  • the image-to-image synthesizer 176 may provide or send the customized artworks to the network-connected display device 104 for the unified television application 130 to display as ambient screens (e.g., the ambient screen 109) on the display 132 of the network-connected display device 104.
  • the mobile computing device 102 may be configured to execute the TV application 110.
  • the mobile computing device 102 may include the mobile computing device display 108 configured to display the UI 114.
  • a user may interact with the UI 114 to set up, control, and interact with the TV application 110.
  • the TV application 110 may display the virtual remote control 138 in the UI 114 allowing the user 101 to interact with and control the network-connected display device 104 and/or the media adapter 107.
  • the mobile computing device 102 may be any type of computing device that includes one or more processors (processor(s) 140), one or more memory devices (memory device(s) 142), and an operating system 144.
  • the mobile computing device 102 may be a smartphone, a tablet, a wearable device, a laptop computer, or a desktop computer.
  • the operating system 144 may be system software that manages computer hardware, software resources, and provides common services for computing programs.
  • the mobile computing device 102 may be a tablet, a smartphone, or a wearable. In these implementations, the operating system 144 may be referred to as a mobile operating system.
  • the mobile operating system may be configured to execute on devices that, in general, include display devices that may be smaller in size than, for example, a display device included in a laptop computer or a desktop computer.
  • the mobile computing device 102 may be a laptop computer.
  • the operating system may be referred to as a laptop or desktop operating system.
  • the operating system 144 may be an operating system designed for a display that is larger in size than that included in a tablet, a smartphone, or a wearable.
  • the media adapter 107 e.g., a casting device, a media streaming device, a media streaming player, a set-top box
  • the media adapter 107 may be interfaced with or connected to the network-connected display device 104.
  • the media adapter 107 may interact with and communicate with the media content providers 160, the server computer 106, and the mobile computing device 102 when providing media content to the network-connected display device 104.
  • the media adapter 107 may be embedded in and/or an integrated part of the network-connected display device 104.
  • the media content providers 160 may include a variety of streaming service and media content sources and service platforms.
  • the media adapter 107 may facilitate providing (e.g., streaming) media content (e.g., streaming video such as movies, TV shows, etc.) from one or more streaming services included in the media content providers 160 to the network-connected display device 104.
  • the media adapter 107 may directly connect to a connector on the network-connected display device 104 by way of connection 165.
  • the media adapter 107 may provide digital video and/or audio to the network- connected display device 104.
  • the media adapter 107 may connect to a high- definition multimedia interface (HDMI) connector included in the network-connected display device 104.
  • HDMI high- definition multimedia interface
  • Examples of the media adapter 107 may include, but are not limited to, a set-top box, a television box, and a streaming media adapter. Atty Docket No.0120-996WO1 [0070]
  • the mobile computing device 102 may connect to or interface with the media adapter 107 by way of a wireless communication link 163b.
  • Wireless communication links 163a-e may be short-range wireless connections such as a Bluetooth connection.
  • wireless communication links 163a-e may be a Wi-Fi (e.g., direct Wi-Fi) connection.
  • the media adapter 107 may be any type of computing device that includes one or more processors (processor(s) 170), one or more memory devices (memory device(s) 172), and an operating system 174.
  • the processor(s) 170 may include a system on a chip (SoC).
  • SoC may include a central processing unit (CPU), a graphic processing unit (GPU), one or more memory interfaces, and one or more input/output interfaces and devices.
  • the operating system 174 may be system software that manages computer hardware, software resources, and provides common services for computing programs.
  • the network-connected display device 104 may include the unified television application 130.
  • the unified television application 130 may keep a record of the interactions of the user with the media content received from the server computer 106.
  • the network- connected display device 104 may send the record of the interactions to the server computer 106 for use in determining media content recommendations for the user.
  • the network-connected display device 104 may be configured to execute the unified television application 130.
  • the network- connected display device 104 may be a smart television.
  • a smart television may be a network-connected television that may connect to media content providers (e.g., media content providers 160) by way of a network (e.g., the network 150).
  • the media content providers may source media content to the smart television.
  • a user may interact with the unified television application 130 to access media content from the media content providers 160.
  • the unified television application 130 may interface with the server computer 106, and specifically with the server-side TV application 116.
  • the unified television application 130 may provide similar functionality to the user as that provided by an application executing on the media adapter 107.
  • the network-connected display device 104 may be configured to connect to the network 150.
  • the network-connected display device 104 is a television (e.g., a smart television (TV)).
  • the network-connected display device 104 may include one or more processors (processor(s) 156), one or more memory devices (memory device(s) 152), and an operating system (OS) 154.
  • the operating system 154 may execute (or assist with executing) the unified television application 130.
  • the operating system 154 may be a browser application.
  • a browser application is a web browser configured to access information on the Internet by way of a network (e.g., the network 150).
  • a browser application may launch one or more browser tabs in the context of one or more browser windows in the browser application.
  • the operating system 154 is a Linux-based operating system configured to execute (or assist with executing) the unified television application 130.
  • the system 100 may include one or more server computers (e.g., the server computer 106) configured to interface with the mobile computing device 102, the media adapter 107, the media content providers 160, and the network-connected display device 104 by way of the network 150.
  • the network 150 may establish a wireless communication link between the network-connected display device 104, the mobile computing device 102, the media adapter 107, the media content providers 160, and the server computer 106.
  • the server computer 106 may include the unified media platform (UMP) 158.
  • the UMP 158 may facilitate the providing of media content items to the network-connected display device 104 as described herein.
  • the server computer 106 may include the server- side TV application 116.
  • the server-side TV application 116 may facilitate providing the media content items for playing on the network-connected display device 104.
  • the mobile computing device 102 may include the mobile computing device display 108.
  • the mobile computing device display 108 is a display device such as a liquid crystal display (LCD), a light-emitting diode (LED) display, an organic light-emitting diode (OLED) display, or an active-matrix organic light-emitting Atty Docket No.0120-996WO1 diode (AMOLED) display.
  • the network-connected display device 104 may include the display 132.
  • the display 132 is a display device such as a liquid crystal display (LCD), a light-emitting diode display (LED) display, a plasma display, a quantum dot light-emitting diode display (QLED) display, or an organic light-emitting diode (OLED) display.
  • LCD liquid crystal display
  • LED light-emitting diode display
  • QLED quantum dot light-emitting diode display
  • OLED organic light-emitting diode
  • the processor(s) 156, the processor(s) 140, the processor(s) 170, and the processor(s) 180 may be formed in a substrate configured to execute one or more machine executable instructions or pieces of software, firmware, or a combination thereof.
  • the processor(s) 156, the processor(s) 140, the processor(s) 170, and the processor(s) 180 may be semiconductor-based.
  • the processor(s) 156, the processor(s) 140, the processor(s) 170, and the processor(s) 180 may include semiconductor material that can perform digital logic.
  • the memory device(s) 152, the memory device(s) 142, the memory device(s) 172, and the memory device(s) 182 may include main memory that stores information in a format that can be read and/or executed by the processor(s) 156, the processor(s) 140, the processor(s) 170, and the processor(s) 180 respectively.
  • the memory device(s) 152, the memory device(s) 142, the memory device(s) 172, and the memory device(s) 182 may include one or more random-access memory (RAM) devices and/or one or more read-only memory (ROM) devices. [0081]
  • the memory device(s) 152, memory device(s) 142, the memory device(s) 172, and the memory device(s) 182 may store applications that, when executed by the processor(s) 156, the processor(s) 140, the processor(s) 170, and the processor(s) 180, respectively, perform operations.
  • the memory device(s) 142 may store the operating system 144 and the TV application 110 that, when executed by the processor(s) 140, may perform operations on the mobile computing device 102.
  • the memory device(s) 152 may store the operating system 154 and the unified television application 130 that, when executed by the processor(s) 156, may perform operations on the network- connected display device 104.
  • the memory device(s) 182 may represent any kind of (or multiple kinds of) memory (e.g., RAM, flash, cache, disk, tape, etc.).
  • the memory device(s) 182 may include external storage, e.g., memory Atty Docket No.0120-996WO1 physically remote from but accessible by the server computer 106.
  • the server computer 106 may include one or more modules, engines, or applications representing specially programmed software.
  • the server computer 106 may include the operating system 184, the server-side TV application 116, the knowledge module 166, the AI module 194 the includes the Gen AI engine 162 and the generative AI model(s) 164, the UMP 158, the images repository 168, the image-to-image synthesizer 176, processor(s) 180, and the memory device(s) 182.
  • the memory device(s) 182 may store the operating system 184, the server-side TV application 116, the knowledge module 166, the AI module 194 including the generative AI engine 162 and the generative AI model(s) 164, the UMP 158, the images repository 168, and the image-to-image synthesizer 176 that, when executed by the processor(s) 180, may perform operations on server computer 106 to implement one or more of the methods and processes described herein.
  • the network 150 may include the Internet and/or other types of data networks, such as a local area network (LAN), a wide area network (WAN), a cellular network, satellite network, or other types of data networks.
  • the network 150 may also include any number of computing devices (e.g., computer, servers, routers, network switches, etc.) that are configured to receive and/or transmit data within the network 150.
  • the network 150 may further include any number of hardwired and/or wireless connections.
  • the network 150 may be, for example, communications networks having one or more types of topologies, including but not limited to the Internet, intranets, local area networks (LANs), cellular networks, Ethernet, Storage Area Networks (SANs), telephone networks, and Bluetooth personal area networks (PAN).
  • LANs local area networks
  • SANs Storage Area Networks
  • PAN personal area networks
  • FIG. 2 is an illustration of an example process 200 for generating a customized artwork (e.g., output image 204) based on an input image 202.
  • the input image 202 can be selected from a carousel of features images (e.g., the featured images carousel 134).
  • the image-to-image synthesizer 176 may perform the process 200.
  • the process 200 includes the image-to-image synthesizer 176 using the ensemble based generative AI model 178 to create or generate the output image 204.
  • the Atty Docket No.0120-996WO1 process 200 may synthesize or collect a dataset for generating the output image 204 based on the input image 202.
  • the image-to-image synthesizer 176 can use the ensemble based generative AI model 178 to extract information and data related to the content of the input image 202.
  • the ensemble based generative AI model 178 may include an image comprehension model 208.
  • the image comprehension model 208 may receive the input image 202 as input and determine an image comprehension keyword token 210.
  • the image comprehension keyword token is a keyword that identifies a key part, component, or object in the input image 202.
  • the image comprehension keyword token 210 may be “frog”.
  • the image comprehension keyword token 210 may be “snowmen”.
  • the image comprehension keyword token 210 may be “cat”.
  • the image comprehension model 208 may be a family of large language models (LLMs) that can recognize and interpret the content of the input image 202.
  • the image comprehension model 208 may determine at least one image comprehension keyword token 210 associated with the input image 202.
  • the image comprehension model 208 may determine additional keyword tokens 206a-e associated with additional information and data for the input image 202.
  • the ensemble based generative AI model 178 may combine the image comprehension keyword token 210 and the keyword tokens 206a-e to form a homologated keyword token 212.
  • the homologated keyword token 212 may be representative of the content of the input image 202.
  • the homologated keyword token 212 may be considered an official keywork token for associating with the input image 202.
  • FIG.3A is an illustration of a first image 300 that includes three snowmen.
  • FIG. 3B is an illustration of a second image 350 that is based on the first image 300. For example, referring to FIGS.
  • the first image 300 may be on a sweater or scarf that belongs to a user (e.g., the user 101).
  • the first image 300 may be included in the images repository 168.
  • the input image 202 may be the first image 300.
  • the input image 202 (the first image 300) may be input to the image comprehension model 208 of the ensemble based generative AI model 178.
  • the image comprehension model 208 may output the image comprehension keyword token 210 (e.g., “snowmen”) in addition to the keyword Atty Docket No.0120-996WO1 tokens 206a-e (e.g., “three”, “wearing”, “hats”, “and”, “scarves”, respectively).
  • the image comprehension keyword token 210 and the keyword tokens 206a-e are formed into a homologated keyword token 212.
  • the image-to-image synthesizer 176 generates the homologated keyword token 212 for use as a basis for generating the output image 204.
  • the homologated keyword token 212 may be input to an image generation model 214.
  • the image generation model 214 may use a text-to-image diffusion model that incorporates a degree of photorealism and a deep level of language understanding to generate the output image 204.
  • the user 101 may request an artistic rendition of the input image 202 (e.g., the first image 300 as shown in FIG.3A). As described herein, the artistic rendition may be based on an artistic style preferred by the user.
  • the Gen AI model 406 may receive images (e.g., first image 420a and second image 420b) as input images from a web crawler 424.
  • the web crawler 424 may provide the images 420a-b that include the same or a similar object (e.g., two images that each include a frog).
  • the Gen AI model 406 may include an image comprehension model 408.
  • the image comprehension model 408 may receive the images Atty Docket No.0120-996WO1 420a-b as input and determine an image comprehension keyword token 410 common to both images 420a-b.
  • the image comprehension model 208 may be a family of large language models (LLMs) that can recognize and interpret the content of the images 420a-b.
  • LLMs large language models
  • the image comprehension model 208 may determine at least one image comprehension keyword token 410 associated with each of the images 420a-b. In addition, or in the alternative, the image comprehension model 208 may determine additional keyword tokens 416a-b associated with additional information and data for the first image 420a and additional keyword tokens 418a- b associated with additional information and data for the second image 420b.
  • the Gen AI model 406 may receive the images 420a-b and determine that there is a common object included in each image (e.g., a frog).
  • the image comprehension model 408 may output the image comprehension keyword token 410 (e.g., “frog”) that identifies a common object between the images 420a-b (e.g., a frog).
  • the image comprehension keyword token 410 for the identified common object between the images 420a-b may be used as a keyword token for each image (e.g., first image keyword token 416a (e.g., “frog”) for the first image 420a, and second image keyword token 418a (e.g., “frog”) for the second image 420b).
  • the image comprehension model 408 may provide additional keyword tokens for each image (e.g., first image additional keyword token(s) 416b for the first image 420a and second image additional keyword token(s) 418b for the second image 420b) that are differentiators between first image 420a and second image 420b.
  • the keyword-based summarizer 412 may generate the keyword-based summaries for each image as shown in Equation 1 for the first image 420a (Sum[Frog1_Image]) and Equation 2 for the second image 420b (Sum[Frog2_Image]).
  • keyword[frog] the first image keyword token 416a, [Keyword 1] ... [Keyword n] are the first image additional keyword token(s) 416b.
  • the Gen AI model 406 may include an image generation model 414.
  • the Gen AI model 406 may use the image generation model 414 to generate an image of the common object as included in the second image 420b (e.g., the frog in the second image 420b) based on the differentiating keyword tokens for the first image 420a (e.g., first image additional keyword token(s) 416b) as the output image 404.
  • the process 200 and the process 400 may provide datasets for output images that have a fixed dimensional output (e.g., a fixed aspect ratio). For example, the dimensions of images for display on the screen of a smart TV are typically in a 16:9 aspect ratio. In addition, or in the alternative, process 200 and process 400 may also provide additional words and/or images that fit into the fixed dimensional output. [0099] In some implementations, large language models (LLMs) may utilize models that include billions of parameters. In addition, or in the alternative, the LLMs may have a latency time that may be limiting for an end user of the system utilizing the model.
  • LLMs large language models
  • Style transfer based generative adversarial networks (GANs) and cyclic GANs may be suited for use by image-to-image synthesizers.
  • the style- transfer based GANs may be used by an image generation model (e.g., the image generation model 214, the image generation model 414) included in an image-to-image synthesizer (e.g., the image-to-image synthesizer 176, the image-to-image synthesizer 476).
  • the image generation model (e.g., the image generation model 214, the image generation model 414) Atty Docket No.0120-996WO1 may be finetuned and refined using intelligent-pruning and reinforcement learning (RL) based model compression techniques.
  • the refining and finetuning of an image generation model may reduce the number of parameters typically used by LLMs (on the order of billions) by approximately 99%. In these implementations, this significant reduction in the number of parameters used by the LLMs can allow the image generation model to run on a smart TV or a low-cost backend server.
  • the image-to- image synthesizer (e.g., the image-to-image synthesizer 176, the image-to-image synthesizer 476) can operate on a network-connected display device (e.g., the network-connected display device 104) such as a smart TV.
  • a user may provide information and data that may be used for preemptive intelligent suggestions for keyword tokens.
  • a user may verbally articulate preferences for output images using, for example, a voice assistant on the smart TV.
  • preferences for output images may be based on past user experience, watch history, regional trends, etc.
  • FIG.5 illustrates a flowchart depicting example operations of generating an image (e.g., a customized artwork) for display in an ambient screen of a television application executing on a display device according to implementations described throughout this disclosure.
  • an image e.g., a customized artwork
  • FIG. 5 illustrates the operations in sequential order, it will be appreciated that this is merely an example, and that additional or alternative operations may be included. Further, operations of FIG. 5 and related operations may be executed in a different order than that shown, or in a parallel or overlapping fashion.
  • the operations may define a computer-implemented method.
  • Operation 510 includes sending, by a server computer to a display device, images for including in a carousel on the display device, the images included in the carousel for use as a basis for generating customized artwork for display in an ambient screen of a television application executing on the display device.
  • Operation 520 includes generating an output image including receiving a selection of an input image from the carousel, extracting image content information and data from the input image, determining keyword tokens for associating with the input image; and applying an ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens.
  • Operation 530 includes sending, by the server computer to the display device, the output image as the customized artwork for display in the ambient screen of the television application executing on the display device.
  • the image-to-image synthesizer 176 may apply the ensemble based generative AI model 178 to an input image (e.g., input image 202) to generate the output image 204.
  • the server computer 106 may send the output image 204 to the network-connected display device 104 for display in the ambient screen 109 of the unified television application 130.
  • the following examples can be combined with one another in any suitable combination. Features and examples described herein with respect to the method can be implemented in non-transitory computer-readable medium and/or by the system, and vice versa.
  • the techniques described herein relate to a method including: sending, by a server computer to a display device, images for including in a carousel on the display device, the images included in the carousel for use as a basis for generating customized artwork for display in an ambient screen of a television application executing on the display device; generating an output image including: receiving a selection of an input image from the carousel; extracting image content information and data from the input image; determining keyword tokens for associating with the input image; and applying an ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens; and sending, by the server computer to the display device, the output image as the customized artwork for display in the ambient screen of the television application executing on the display device.
  • the techniques described herein relate to a method, wherein extracting image content information and data from the input image includes extracting the image content information and data from the input image using generative artificial intelligence.
  • the techniques described herein relate to a method, wherein generating the output image further includes applying a generative artificial intelligence model that includes an image comprehension model and an image generation model.
  • the techniques described herein relate to a method, wherein the ensemble of style transfer networks and generative adversarial networks are included in the image generation model.
  • the techniques described herein relate to a method, wherein extracting image content information and data from the input image includes extracting the image content information and data from the input image by the image comprehension model.
  • the techniques described herein relate to a method, wherein the keyword tokens include an image comprehension keyword token; and wherein the image comprehension model determines the image comprehension keyword token for the input image based on the image content information and data for the input image.
  • the techniques described herein relate to a method, wherein the image generation model generates the output image as an artistic rendering of the input image.
  • the techniques described herein relate to a method, wherein the keyword tokens include an image comprehension keyword token and at least one additional keyword token; and wherein generating the output image further includes generating a homologated keyword token based on the image comprehension keyword token and the at least one additional keyword token.
  • the techniques described herein relate to a method, wherein applying the ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens includes Atty Docket No.0120-996WO1 applying the ensemble of style transfer networks and generative adversarial networks on the homologated keyword token.
  • the techniques described herein relate to a method, wherein the output image is generated by an image-to-image synthesizer.
  • the techniques described herein relate to a non-transitory computer-readable medium storing executable instructions that when executed by at least one processor of a server computer cause the at least one processor to execute operations, the operations including: sending, to a display device, images for including in a carousel on the display device, the images included in the carousel for use as a basis for generating customized artwork for display in an ambient screen of a television application executing on the display device; generating an output image including: receiving a selection of an input image from the carousel; extracting image content information and data from the input image; determining keyword tokens for associating with the input image; and applying an ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens; and sending, to the display device, the output image as the customized artwork for display in the ambient screen of
  • the techniques described herein relate to a non-transitory computer-readable medium, wherein extracting image content information and data from the input image includes extracting the image content information and data from the input image using generative artificial intelligence.
  • the techniques described herein relate to a non-transitory computer-readable medium, wherein generating the output image further includes applying a generative artificial intelligence model that includes an image comprehension model and an image generation model.
  • the techniques described herein relate to a non-transitory computer-readable medium, wherein the ensemble of style transfer networks and generative adversarial networks are included in the image generation model.
  • the techniques described herein relate to a non-transitory computer-readable medium, wherein extracting image content information and data from the Atty Docket No.0120-996WO1 input image includes extracting the image content information and data from the input image by the image comprehension model.
  • the techniques described herein relate to a non-transitory computer-readable medium, wherein the keyword tokens include an image comprehension keyword token; and wherein the image comprehension model determines the image comprehension keyword token for the input image based on the image content information and data for the input image.
  • the techniques described herein relate to a non-transitory computer-readable medium, wherein the image generation model generates the output image as an artistic rendering of the input image.
  • the techniques described herein relate to a non-transitory computer-readable medium, wherein the keyword tokens include an image comprehension keyword token and at least one additional keyword token; and wherein generating the output image further includes generating a homologated keyword token based on the image comprehension keyword token and the at least one additional keyword token.
  • the techniques described herein relate to a non-transitory computer-readable medium, wherein applying the ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens includes applying the ensemble of style transfer networks and generative adversarial networks on the homologated keyword token.
  • the techniques described herein relate to a non-transitory computer-readable medium, wherein the output image is generated by an image-to-image synthesizer.
  • the techniques described herein relate to a system including: at least one processor; and a non-transitory computer-readable medium storing instructions that when executed by the at least one processor cause the system to: send, from a server computer to a display device, images for including in a carousel on the display device, the images included in the carousel for use as a basis for generating customized artwork for display in an ambient screen of a television application executing on the display device; generate an output image including: receiving a selection of an input image from the carousel; extracting image content information and data from the input image; determining Atty Docket No.0120-996WO1 keyword tokens for associating with the input image; and applying an ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens;
  • the techniques described herein relate to a system, wherein extracting image content information and data from the input image includes extracting the image content information and data from the input image using generative artificial intelligence.
  • the techniques described herein relate to a system, wherein generating the output image further includes applying a generative artificial intelligence model that includes an image comprehension model and an image generation model.
  • the techniques described herein relate to a system, wherein the ensemble of style transfer networks and generative adversarial networks are included in the image generation model.
  • the techniques described herein relate to a system, wherein extracting image content information and data from the input image includes extracting the image content information and data from the input image by the image comprehension model.
  • the techniques described herein relate to a system, wherein the keyword tokens include an image comprehension keyword token; wherein the image comprehension model determines the image comprehension keyword token for the input image based on the image content information and data for the input image. [0092] In some examples, the techniques described herein relate to a system, wherein the image generation model generates the output image as an artistic rendering of the input image. [0093] In some examples, the techniques described herein relate to a system, wherein the keyword tokens include an image comprehension keyword token and at least one additional keyword token; and wherein generating the output image further includes generating a homologated keyword token based on the image comprehension keyword token and the at least one additional keyword token.
  • the techniques described herein relate to a system, wherein applying the ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens includes applying the ensemble of style transfer networks and generative adversarial networks on the homologated keyword token.
  • the techniques described herein relate to a system, wherein the output image is generated by an image-to-image synthesizer.
  • ASICs application specific integrated circuits
  • These various implementations can include implementation in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.
  • programmable processor which may be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.
  • These computer programs also known as programs, software, software applications or code
  • machine-readable medium refers to any computer program product, apparatus and/or device (e.g., magnetic discs, optical disks, memory, Programmable Logic Devices (PLDs)) used to provide machine instructions and/or data to a programmable processor, including a non-transitory machine-readable medium that receives machine instructions as a machine-readable signal.
  • machine-readable signal refers to any signal used to provide machine instructions and/or data to a programmable processor.
  • a computer program product comprising computer-executable instructions which, when executed by at least one computing apparatus, cause the at least one computing apparatus to perform the method described herein may also be provided.
  • the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a Atty Docket No.0120-996WO1 keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer.
  • a display device e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor
  • a pointing device e.g., a mouse or a trackball
  • Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.
  • the systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a client computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components.
  • the components of the system can be interconnected by any form or non-transitory medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (“LAN”), a wide area network (“WAN”), and the Internet.
  • LAN local area network
  • WAN wide area network
  • the Internet the global information network
  • the computing system can include clients and servers.
  • a client and server are generally remote from each other and typically interact through a communication network.
  • the relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.
  • the singular forms "a,” “an” and “the” do not exclude the plural reference unless the context clearly dictates otherwise.
  • conjunctions such as “and,” “or,” and “and/or” are inclusive unless the context clearly dictates otherwise.
  • “A and/or B” includes A alone, B alone, and A with B.
  • connecting lines or connectors shown in the various figures presented are intended to represent example functional relationships and/or physical or logical couplings between the various elements.
  • a user may be provided with controls allowing the user to make an election as to both if and when systems, programs, or features described herein may enable collection of user information (e.g., a user’s preferences, a user’s current location, a user’s credentials, etc.), and if the user is sent content or communications from a server.
  • user information e.g., a user’s preferences, a user’s current location, a user’s credentials, etc.
  • certain data may be treated in one or more ways before it is stored or used, so that personally identifiable information is removed.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Signal Processing (AREA)
  • Multimedia (AREA)
  • Computing Systems (AREA)
  • General Physics & Mathematics (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computational Linguistics (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Probability & Statistics with Applications (AREA)
  • Computer Graphics (AREA)
  • Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)

Abstract

According to an aspect, a method may send, by a server computer to a display device, images for including in a carousel on the display device, the images included in the carousel for use as a basis for generating customized artwork for display in an ambient screen of a television application executing on the display device. A method may generate an output image including receiving a selection of an input image from the carousel, extracting image content information and data from the input image, determining keyword tokens for associating with the input image, and applying an ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens. A method may send, by the server computer to the display device, the output image as the customized artwork for display in the ambient screen of the television application executing on the display device.

Description

Atty Docket No.0120-996WO1 CONTENT FOR DISPLAYING IN AN AMBIENT MODE SCREEN FOR A TELEVISION APPLICATION BACKGROUND [0001] A television (TV) application may present various types of media content of interest to a user on a smart TV. During times when a user is not interacting with the TV application or watching media content on the smart TV, the TV application may display an ambient screen on the smart TV. In some situations, the ambient screen may be a blank screen. In some situations, the ambient screen may display an image that will not cause any additional wear, burn-in, or damage to the display of the smart TV. SUMMARY [0002] A smart television in a room of a home of a user may be the centerpiece of the room due to its large size and prominent placement in the room. The smart TV may provide an ambient screen or a home screen when the user is not interacting with or watching media content on the smart TV. For example, when the user turns on the smart TV, the smart TV may provide the ambient screen until the user selects or clicks a button on a remote control for the smart TV. In some implementations, the ambient screen may be a blank screen (no image is displayed on the screen of the smart TV). In some implementations, a TV application running on the smart TV may display images in an iterative manner from a photo collection of the user. [0003] Because the smart TV may be prominently placed in a room, the user may want to have the smart TV display an image on the ambient screen. Some users may want to have the smart TV display an interesting, artistic looking image on the ambient screen making the smart TV appear to be a piece of artwork in the room. In addition, or in the alternative, the TV application may provide additional information on the ambient screen that may include, but is not limited to, a time and date, current weather conditions, a news headline, a calendar reminder, and one or more minimalistic advertisements. Each information entry on the ambient screen may be selected or clicked on by the user interacting with a remote control for the smart TV. Once selected, each entry may direct the user to a web site or application on the smart TV. For example, selecting the news headline may direct the user to an application on the smart TV that provides news-based streaming media Atty Docket No.0120-996WO1 content. For example, selecting a minimalistic advertisement may direct the user to a web site to obtain more information related to the content of the advertisement (e.g., how to purchase an item). [0004] In some aspects, the techniques described herein relate to a method including: sending, by a server computer to a display device, images for including in a carousel on the display device, the images included in the carousel for use as a basis for generating customized artwork for display in an ambient screen of a television application executing on the display device; generating an output image including: receiving a selection of an input image from the carousel; extracting image content information and data from the input image; determining keyword tokens for associating with the input image; and applying an ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens; and sending, by the server computer to the display device, the output image as the customized artwork for display in the ambient screen of the television application executing on the display device. [0005] In some aspects, the techniques described herein relate to a method, wherein extracting image content information and data from the input image includes extracting the image content information and data from the input image using generative artificial intelligence. [0006] In some aspects, the techniques described herein relate to a method, wherein generating the output image further includes applying a generative artificial intelligence model that includes an image comprehension model and an image generation model. [0007] In some aspects, the techniques described herein relate to a method, wherein the ensemble of style transfer networks and generative adversarial networks are included in the image generation model. [0008] In some aspects, the techniques described herein relate to a method, wherein extracting image content information and data from the input image includes extracting the image content information and data from the input image by the image comprehension model. [0009] In some aspects, the techniques described herein relate to a method, wherein the keyword tokens include an image comprehension keyword token; and wherein the image Atty Docket No.0120-996WO1 comprehension model determines the image comprehension keyword token for the input image based on the image content information and data for the input image. [0010] In some aspects, the techniques described herein relate to a method, wherein the image generation model generates the output image as an artistic rendering of the input image. [0011] In some aspects, the techniques described herein relate to a method, wherein the keyword tokens include an image comprehension keyword token and at least one additional keyword token; and wherein generating the output image further includes generating a homologated keyword token based on the image comprehension keyword token and the at least one additional keyword token. [0012] In some aspects, the techniques described herein relate to a method, wherein applying the ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens includes applying the ensemble of style transfer networks and generative adversarial networks on the homologated keyword token. [0013] In some aspects, the techniques described herein relate to a method, wherein the output image is generated by an image-to-image synthesizer. [0014] In some aspects, the techniques described herein relate to a non-transitory computer-readable medium storing executable instructions that when executed by at least one processor of a server computer cause the at least one processor to execute operations, the operations including: sending, to a display device, images for including in a carousel on the display device, the images included in the carousel for use as a basis for generating customized artwork for display in an ambient screen of a television application executing on the display device; generating an output image including: receiving a selection of an input image from the carousel; extracting image content information and data from the input image; determining keyword tokens for associating with the input image; and applying an ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens; and sending, to the display device, the output image as the customized artwork for display in the ambient screen of the television application executing on the display device. Atty Docket No.0120-996WO1 [0015] In some aspects, the techniques described herein relate to a non-transitory computer-readable medium, wherein extracting image content information and data from the input image includes extracting the image content information and data from the input image using generative artificial intelligence. [0016] In some aspects, the techniques described herein relate to a non-transitory computer-readable medium, wherein generating the output image further includes applying a generative artificial intelligence model that includes an image comprehension model and an image generation model. [0017] In some aspects, the techniques described herein relate to a non-transitory computer-readable medium, wherein the ensemble of style transfer networks and generative adversarial networks are included in the image generation model. [0018] In some aspects, the techniques described herein relate to a non-transitory computer-readable medium, wherein extracting image content information and data from the input image includes extracting the image content information and data from the input image by the image comprehension model. [0019] In some aspects, the techniques described herein relate to a non-transitory computer-readable medium, wherein the keyword tokens include an image comprehension keyword token; and wherein the image comprehension model determines the image comprehension keyword token for the input image based on the image content information and data for the input image. [0020] In some aspects, the techniques described herein relate to a non-transitory computer-readable medium, wherein the image generation model generates the output image as an artistic rendering of the input image. [0021] In some aspects, the techniques described herein relate to a non-transitory computer-readable medium, wherein the keyword tokens include an image comprehension keyword token and at least one additional keyword token; and wherein generating the output image further includes generating a homologated keyword token based on the image comprehension keyword token and the at least one additional keyword token. [0022] In some aspects, the techniques described herein relate to a non-transitory computer-readable medium, wherein applying the ensemble of style transfer networks and generative adversarial networks on the image content information and data using the Atty Docket No.0120-996WO1 keyword tokens includes applying the ensemble of style transfer networks and generative adversarial networks on the homologated keyword token. [0023] In some aspects, the techniques described herein relate to a non-transitory computer-readable medium, wherein the output image is generated by an image-to-image synthesizer. [0024] In some aspects, the techniques described herein relate to a system including: at least one processor; and a non-transitory computer-readable medium storing instructions that when executed by the at least one processor cause the system to: send, from a server computer to a display device, images for including in a carousel on the display device, the images included in the carousel for use as a basis for generating customized artwork for display in an ambient screen of a television application executing on the display device; generate an output image including: receiving a selection of an input image from the carousel; extracting image content information and data from the input image; determining keyword tokens for associating with the input image; and applying an ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens; and send, from the server computer to the display device, the output image as the customized artwork for display in the ambient screen of the television application executing on the display device. [0025] In some aspects, the techniques described herein relate to a system, wherein extracting image content information and data from the input image includes extracting the image content information and data from the input image using generative artificial intelligence. [0026] In some aspects, the techniques described herein relate to a system, wherein generating the output image further includes applying a generative artificial intelligence model that includes an image comprehension model and an image generation model. [0027] In some aspects, the techniques described herein relate to a system, wherein the ensemble of style transfer networks and generative adversarial networks are included in the image generation model. [0028] In some aspects, the techniques described herein relate to a system, wherein extracting image content information and data from the input image includes extracting the Atty Docket No.0120-996WO1 image content information and data from the input image by the image comprehension model. [0029] In some aspects, the techniques described herein relate to a system, wherein the keyword tokens include an image comprehension keyword token; wherein the image comprehension model determines the image comprehension keyword token for the input image based on the image content information and data for the input image. [0030] In some aspects, the techniques described herein relate to a system, wherein the image generation model generates the output image as an artistic rendering of the input image. [0031] In some aspects, the techniques described herein relate to a system, wherein the keyword tokens include an image comprehension keyword token and at least one additional keyword token; and wherein generating the output image further includes generating a homologated keyword token based on the image comprehension keyword token and the at least one additional keyword token. [0032] In some aspects, the techniques described herein relate to a system, wherein applying the ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens includes applying the ensemble of style transfer networks and generative adversarial networks on the homologated keyword token. [0033] In some aspects, the techniques described herein relate to a system, wherein the output image is generated by an image-to-image synthesizer. [0034] The details of one or more implementations are set forth in the accompanying drawings and the description below. Other features will be apparent from the description and drawings, and from the claims. BRIEF DESCRIPTION OF THE DRAWINGS [0035] FIG.1A illustrates an example of a user interacting with a network-connected display device and a media adapter, according to implementations described throughout this disclosure. [0036] FIG. 1B illustrates an example system for selecting, generating, and displaying an artistic-looking image on an ambient screen of a smart TV according to implementations described throughout this disclosure. Atty Docket No.0120-996WO1 [0037] FIG. 1C illustrates an example illustration of a process flow for selecting, generating, and displaying a customized artwork on an ambient screen of a smart TV according to implementations described throughout this disclosure. [0038] FIG. 2 is an illustration of an example process for generating a customized artwork based on an input image according to implementations described throughout this disclosure. [0039] FIG.3A is an illustration of a first image that includes three snowmen. [0040] FIG.3B is an illustration of a second image that is based on and is an artistic rendering of a first image. [0041] FIG.3C is an illustration of an ambient screen of a smart TV displaying an output image 204 as a customized artwork. [0042] FIG. 4 is an illustration of another example process for generating and displaying a customized artwork on an ambient screen of a smart TV according to implementations described throughout this disclosure. [0043] FIG.5 illustrates a flowchart depicting example operations of generating an image (e.g., a customized artwork) for display in an ambient screen of a television application executing on a display device according to implementations described throughout this disclosure. DETAILED DESCRIPTION [0044] A user may place a smart television (TV) in a prominent spot in a room making the smart TV a focal point of the room. The user may want the smart TV, when not streaming media content for viewing by the user, to display an image on the ambient or home screen of the smart TV. A television application running on a smart television (TV) may facilitate the displaying of the ambient screen image. The ambient screen image may be a user selected image that may also include one or more entries associated with other information that may be of interest to the user. The additional entries may be informative, selectable entries such as a current time and date, current weather conditions, a current news headline, a current calendar entry, and one or more minimalistic advertisements. The user may want the image to be a pleasant, artistic looking image. [0045] The smart TV may provide the ambient screen when a user is not interacting with or watching media content on the smart TV, for example, when the smart TV is first Atty Docket No.0120-996WO1 turned on. The smart TV may provide and, in addition, update the content of the ambient screen until the user selects or clicks a button on a remote control for the smart TV that initiates another action on the smart TV. For example, selecting or clicking a guide button on the remote control may exit the ambient screen mode and the TV application may then display a channel guide to the user. In another example, the user may use the remote control to select one of the entries on the ambient screen that then directs the user to web sites or applications on the smart TV. For example, selecting a news headline may direct the user to an application on the smart TV that provides streaming news media content. For example, selecting a minimalistic advertisement may direct the user to a web site to obtain more information for a product or service covered by the advertisement. [0046] In some implementations, a server-side TV application may facilitate the selection, generation, and displaying of an image on the ambient screen of the smart TV. The image may be an artistic-looking image. In some implementations, an image-to-image synthesizer may utilize an ensemble based generative AI model that uses style transfer generative adversarial networks (GANS) along with generative artificial intelligence (AI) to create or generate a customized artwork based on particular image content for an image. The image-to-image synthesizer may utilize Generative Adversarial Networks (GANs) when generating the customized artwork. For example, GANs are a type of machine learning algorithm that generates new, previously unseen data that is similar to existing data. The fields of art and synthetic media may use GANs to create new and unique works of art and synthetic media that may be based on existing artworks. For example, style transfer is a technique common in the technical areas of computer vision and graphics that involves generating a new image by combining the content of one image with the style of another image. Style transfer can be used when generating a custom artwork from an image by combining the image with the style of a selected work of art. The use of such an ensemble based generative AI model can allow for the generating of an image more quickly, efficiently, and with fewer resources than previous approaches. [0047] For example, the customized artwork may be based on an image included in an image repository. The image repository may include photos, stock images, and other types of images selected by or associated with the user. In some implementations, the server-side TV application may provide or send images used as a basis for the customized artworks to a Atty Docket No.0120-996WO1 carousel of featured images included in the smart TV. A television (TV) application executing on the smart TV may be configured to display a plurality of customized artworks generated from images included in the features carousel of images on an ambient or home screen of the TV application. The TV application may display the customized artworks sequentially according to a temporal order of the images in the featured carousel of images. The TV application may temporally organize the customized artworks as a carousel, smoothly transitioning from displaying one customized artwork after another customized artwork. In some implementations, a mixer may automatically mix the images included in the carousel of featured images sequentially according to the temporal order. [0048] In some implementations, an image-to-image synthesizer executing on the smart TV may generate the customized artwork on the smart TV utilizing the computing resources of the smart TV. In these implementations, the server-side TV application may interface with the image-to-image synthesizer on the smart TV to provide or send image from an image repository to the image-to-image synthesizer for use in generating a customized artwork for the ambient screen of the smart TV. [0049] At least one technical problem involves how to generate an image for display on an ambient or home screen of a smart TV. As described herein, at least one technical solution to the technical problem generates an image for display based on images included in an image repository for a user. The image is generated by the technical process of applying an ensemble of style transfer networks and generative adversarial networks on the image content information and data using keyword tokens. One implementation of the technical solution described herein uses generative artificial intelligence to extract image content from the featured carousel of images and an image-to-image synthesizer that utilizes an ensemble based generative AI model that uses style transfer generative adversarial networks (GANS) to create or generate a customized artwork. At least one technical effect may be providing an alternative way to generate an image for display on an ambient or home screen of a smart TV. [0050] This approach can further provide a pleasant, artistic-looking image for display on an ambient or home screen of a smart TV based on images included in an image repository for a user. This approach may also enhance monetization and a click-to-convert ratio for the carousel of featured images, as well as the portraying of customized artworks Atty Docket No.0120-996WO1 on the ambient screen of the smart TV which may be the centerpiece of a household of a user. The use of a carousel of images may reduce pixel burn-in and damage or wear to the display. [0051] FIG. 1A illustrates an example of a user 101 interacting with a network- connected display device 104 and a media adapter 107 according to implementations described throughout this disclosure. FIG. 1B illustrates an example system 100 for selecting, generating, and displaying an artistic-looking image (e.g., customized artwork 113) on an ambient screen (e.g., home or ambient screen 109) of a smart TV (e.g., the network-connected display device 104) according to implementations described throughout this disclosure. FIG.1C illustrates an example illustration of a process flow 190 for selecting, generating, and displaying a customized artwork (e.g., customized artwork 113) on an ambient screen (e.g., home or ambient screen 109) of a smart TV (e.g., the network- connected display device 104) according to implementations described throughout this disclosure. [0052] Referring to FIGS. 1A-B, the user 101 may enter a room 115 that includes the network-connected display device 104. The user 101 interacting with a remote control device 105 for the network-connected display device 104 may turn the network-connected display device 104 on. Once powered on, the network-connected display device 104 may display the ambient screen 109. For example, the ambient screen 109 may include the customized artwork 113 and entries 111a-d. Entry 111a shows the current date and time. For example, the user 101 may select or click on the entry 111a. In response to the selection of the entry 111a, the network-connected display device 104 may run a date and time application on the network-connected display device 104. Entry 111b shows the current temperature and weather for a location of the user 101. For example, the user 101 may select or click on the entry 111b. In response to the selection of the entry 111b, the network- connected display device 104 may run a weather application on the network-connected display device 104. Entry 111c shows a recent news headline. For example, the user 101 may select or click on the entry 111c. In response to the selection of the entry 111c, the network-connected display device 104 may run a streaming news media application on the network-connected display device 104. Entry 111d shows a minimalistic advertisement for a product. For example, the user 101 may select or click on the entry 111d. In response to Atty Docket No.0120-996WO1 the selection of entry 111d, the network-connected display device 104 may run a web browser application that may display a web site for the product as advertised in the entry 111d. The date and time application, the weather application, the streaming news media application, and the web browser application may be included in television (TV) application(s) 136 on the network-connected display device 104. For example, the application(s) 136 may include media content provider applications, web browser applications, and other applications that may be installed on the network-connected display device 104 and accessed by the user 101. In some implementations, the network-connected display device 104 may access one or more applications by way of a network 150. The network-connected display device 104 may provide the user 101 with information and data provided by the applications to the user 101, for example, in a user interface 112 on the display 132 of the network-connected display device 104. [0053] Referring to FIGS. 1A-B, the network-connected display device 104 may communicate with a server computer 106 and media content providers 160 by way of the network 150. The media content providers 160, the network-connected display device 104, the server computer 106, and a mobile computing device 102 may interact with and communicate with one other by way of the network 150. In some implementations, the mobile computing device 102 may interface or connect to the media adapter 107 and/or the network-connected display device 104 by way of a wireless communication link that may be a short-range wireless connection such as, for example a Bluetooth connection or a Wi- Fi (e.g., direct Wi-Fi) connection. [0054] In some implementations, referring to FIGS.1A-B, the user 101 may connect to and interact with a media adapter (e.g., the media adapter 107) by way of a network- connected display device (e.g., the network-connected display device 104) using a server- side television (TV) application (e.g., the server-side TV application 116) installed on a server computer (e.g., the server computer 106). The media adapter 107 may be connected or interfaced to the network-connected display device 104. The network-connected display device 104 may be communicatively coupled or connected to the server computer 106 by way of the network 150. A unified media platform (UMP) 158 may provide or serve media content items from the media content providers 160 to the network-connected display device 104 by way of the media adapter 107. The network-connected display device 104 may Atty Docket No.0120-996WO1 execute a unified television application 130 that may interface with the server-side TV application 116 by way of the media adapter 107. The server-side TV application 116 may interface with the unified television application 130 to facilitate providing the customized artwork 113 for the ambient screen 109 of the network-connected display device 104 when the network-connected display device 104 is turned on by the user 101. [0055] In some implementations, referring to FIGS.1A-B, the user 101 may interact with a network-connected display device (e.g., the network-connected display device 104) using a remote control device (e.g., the remote control device 105). In some implementations, a television (TV) application 110 may render a virtual remote control 138 in a user interface (e.g., UI 114) on a display (e.g., a mobile computing device display 108) on the mobile computing device 102. The virtual remote control 138 may allow the mobile computing device 102 to act as a remote control for the network-connected display device 104. The TV application 110 may render the virtual remote control 138 for use with the network-connected display device 104. [0056] The user 101 may interact with the remote control device 105 and/or the virtual remote control 138 when selecting media content for viewing on the network- connected display device 104. In addition, or in the alternative, the user 101 may interact with remote control device 105 and/or the virtual remote control 138 to select or click on and launch applications included in the TV application(s) 136. In addition, or the alternative, the user 101 may interact with the remote control device 105 and/or the virtual remote control 138 to select or click on the customized artwork 113 displayed on the ambient screen 109 to lead the user 101 into a featured images carousel 134 that includes the content or image used as the basis for the customized artwork 113. [0057] In some implementations, referring to FIGS.1A-B, the user 101 may connect to and interact with a media adapter (e.g., the media adapter 107) using a TV application (e.g., the television (TV) application 110) installed on a mobile computing device (e.g., the mobile computing device 102). In some implementations, a user (e.g., the user 101) may connect to and interact with a media adapter (e.g., the media adapter 107) using a media adapter remote control device (e.g., media adapter remote control device 103). In some implementations, the TV application 110 may render the virtual remote control 138 for use with the media adapter 107. The virtual remote control 138 may allow the mobile computing Atty Docket No.0120-996WO1 device 102 to act as a remote control for the media adapter 107. The user (e.g., the user 101) may interact with the virtual remote control 138 and/or the media adapter remote control device 103 when interacting with the media adapter 107. [0058] The network-connected display device 104 may execute the unified television application 130. The unified television application 130 may interface with a server-side television (TV) application 116. The UMP 158 may interface with the unified television application 130 executing on the network-connected display device 104 and the media content providers 160 to provide media content to the network-connected display device 104. [0059] The server computer 106 may include a knowledge module 166. The knowledge module 166 may include information associated with media content items provided by the media content providers 160. In some implementations, the knowledge module 166 may generate media content recommendations for associating with an account of a user based, in part, on a multi-dimensional user activity characteristic associated with the account of the user and the information associated with media content items provided by the media content providers 160. The user activity characteristic associated with the account of the user may be obtained from a plurality of information sources that may include, but are not limited to, a search engine, a mapping application, and an online retailer. The information sources may provide activity data related to activities of the account of the user by way of a respective software program or application. [0060] In some implementations, the unified television application 130 may interface with the knowledge module 166 to provide information and data related to the past activities of the user when interacting with the unified television application 130, the viewing history of the user, and/or the popularity of media content items of a type, classification, category, group or genre. The knowledge module 166 may help the unified television application 130 identify media content that may be useful and of interest to the user. The unified television application 130 interfacing with the knowledge module 166 may curate or provide media content recommendations based on the past activities of the user when interacting with the unified television application 130, the viewing history of the user, and/or the popularity of media content items of a certain type, classification, category, group or genre. The unified television application 130 may provide image(s) related to or associated Atty Docket No.0120-996WO1 with the recommended media content to an images repository 168. In addition, or in the alternative, the images repository 168 may include images associated with a user (e.g., the user 101) such as photographs and image files. Referring to FIGS.1A-C, the server-side TV application 116 may facilitate providing or sending images from the images repository 168 to the featured images carousel 134 for use as home screens for a smart TV (e.g., the network-connected display device 104). [0061] The image-to-image synthesizer 176 may receive images from the images repository 168. The image-to-image synthesizer 176 may use an ensemble based generative AI model 178 included in generative artificial intelligence (Gen AI) model(s) 164 to extract information and data associated with the image (image content) from the images as a basis for generating artworks. For example, the ensemble based generative AI model 178 may utilize generative artificial intelligence and style transfer based Generative Adversarial Networks (GANs) to extract content from images, to generate new data based on and that is similar to existing data for an artwork, and to create or generate a new image (e.g., a customized artwork) by combining the extracted image content of an image of the user (e.g., an image from the images repository 168) with the new data. The image-to-image synthesizer 176 may use GANs to create new and unique works of art and synthetic media that may be based on existing artworks. Style transfer techniques may generate a new image by combining the content of one image with the style of another image. Combining the use of GANS with style transfer techniques may generate or create customized artworks for a user by combining an image selected by a user with the style of a work of art that is also selected by the user. In addition, a technical benefit may be generating or creating the customized artworks on the smart TV itself by implementing the image-to-image synthesizer on the smart TV (e.g., the network-connected display device 104). This may result in a quicker more efficient way of generating or creating the customized artworks. [0062] The unified television application 130 may present a user interface in the user interface 112 that allows the user (e.g., the user 101) to select one or more images from the images repository 168 that the user would like to use in a carousel of featured images for use as a basis for generating customized artworks for presenting in the ambient screen 109 of the network-connected display device 104. The selected image may be provided as an input image. The input image may be sent from the display device to the server computer. Atty Docket No.0120-996WO1 [0063] In some implementations, referring to FIGS. 1A-C and 2, a customized artwork (e.g., the customized artwork 113) may be based on an image included in the images repository 168 and sent or provided to the featured images carousel 134. The image repository may include photos, stock images, and other types of images selected by or associated with the user. In some implementations, the server-side TV application 116 may provide or send images used as a basis for the customized artworks to the featured images carousel 134 included in the network-connected display device 104. The unified television application 130 executing on the network-connected display device 104 may be configured to display multiple customized artworks based on images included in the featured images carousel 134 on the ambient screen 109. The unified television application 130 may cause the generation and display of the customized artworks sequentially according to a temporal order of the images included in the featured images carousel 134. The unified television application 130 may temporally organize the images and as such the customized artworks as a carousel, smoothly transitioning from the displaying of one customized artwork to another customized artwork. In some implementations, the unified television application 130 may automatically mix the images included in the featured images carousel 134 sequentially according to the temporal order. Transitioning between customized artwork in a round robin fashion by accessing a carousel of featured images may reduce pixel burn-in and damage or wear to the display. [0064] The unified television application 130 may present a user interface in the user interface 112 that allows a user (e.g., the user 101) to specify an artwork as the basis for the image-to-image synthesizing. In some implementations, the user may select an artwork by choosing from names of artworks by entering the name of an artwork in a text entry field. In some implementations, the user may select an artwork for a list of artworks presented in a dropdown menu. The names of artworks included for selection in the dropdown menu may be based on information about preferences of the user as provided by the knowledge module 166 (e.g., the user likes contemporary home furnishings and music so the artworks may be artworks by contemporary artists). In some implementations, the user interface 112 may allow the user to enter the name of an artist (e.g., Degas) and then the user interface may present the user with a selection of artworks by the artist in a dropdown menu for the user to choose from. In some implementations, the user may specify a different artwork for use in Atty Docket No.0120-996WO1 generating a customized artwork for each selected image from the images repository 168. In some implementations, the user may specify a single artwork for use in creating a customized artwork for all of the user selected images from the images repository 168. [0065] The server-side TV application 116 and the image-to-image synthesizer 176 may interface with the artificial intelligence (AI) module 194 that includes the generative artificial intelligence (Gen AI) model(s) 164 and a generative artificial intelligence (Gen AI) engine 162. The Gen AI model(s) 164 may be machine learning trained models (e.g., the ensemble based generative AI model 178 described herein) for use by the Gen AI engine 162. The Gen AI engine 162 can use generative artificial intelligence for extracting content from images included in the images repository 168. The image-to-image synthesizer 176 may use the extracted content as a basis for generating or creating customized artwork for displaying as ambient screens (e.g., the ambient screen 109) on the network-connected display device 104 (e.g., a smart TV). The image-to-image synthesizer 176 may interface with the Gen AI engine 162 to use the ensemble based generative AI model 178 on the extracted content from the images by applying the style transfer based GANs to the images to generate customized artworks based on the images. The image-to-image synthesizer 176 may provide or send the customized artworks to the network-connected display device 104 for the unified television application 130 to display as ambient screens (e.g., the ambient screen 109) on the display 132 of the network-connected display device 104. [0066] The mobile computing device 102 may be configured to execute the TV application 110. The mobile computing device 102 may include the mobile computing device display 108 configured to display the UI 114. A user may interact with the UI 114 to set up, control, and interact with the TV application 110. In some implementations, as described, the TV application 110 may display the virtual remote control 138 in the UI 114 allowing the user 101 to interact with and control the network-connected display device 104 and/or the media adapter 107. [0067] The mobile computing device 102 may be any type of computing device that includes one or more processors (processor(s) 140), one or more memory devices (memory device(s) 142), and an operating system 144. The mobile computing device 102 may be a smartphone, a tablet, a wearable device, a laptop computer, or a desktop computer. In some Atty Docket No.0120-996WO1 implementations, the operating system 144 may be system software that manages computer hardware, software resources, and provides common services for computing programs. [0068] In some implementations, the mobile computing device 102 may be a tablet, a smartphone, or a wearable. In these implementations, the operating system 144 may be referred to as a mobile operating system. The mobile operating system may be configured to execute on devices that, in general, include display devices that may be smaller in size than, for example, a display device included in a laptop computer or a desktop computer. In some implementations, the mobile computing device 102 may be a laptop computer. In these implementations, the operating system may be referred to as a laptop or desktop operating system. In these implementations, the operating system 144 may be an operating system designed for a display that is larger in size than that included in a tablet, a smartphone, or a wearable. [0069] In some implementations, the media adapter 107 (e.g., a casting device, a media streaming device, a media streaming player, a set-top box) may be interfaced with or connected to the network-connected display device 104. The media adapter 107 may interact with and communicate with the media content providers 160, the server computer 106, and the mobile computing device 102 when providing media content to the network-connected display device 104. In some implementations, the media adapter 107 may be embedded in and/or an integrated part of the network-connected display device 104. [0028] The media content providers 160 may include a variety of streaming service and media content sources and service platforms. The media adapter 107 may facilitate providing (e.g., streaming) media content (e.g., streaming video such as movies, TV shows, etc.) from one or more streaming services included in the media content providers 160 to the network-connected display device 104. For example, the media adapter 107 may directly connect to a connector on the network-connected display device 104 by way of connection 165. The media adapter 107 may provide digital video and/or audio to the network- connected display device 104. For example, the media adapter 107 may connect to a high- definition multimedia interface (HDMI) connector included in the network-connected display device 104. Examples of the media adapter 107 may include, but are not limited to, a set-top box, a television box, and a streaming media adapter. Atty Docket No.0120-996WO1 [0070] In some implementations, the mobile computing device 102 may connect to or interface with the media adapter 107 by way of a wireless communication link 163b. Wireless communication links 163a-e may be short-range wireless connections such as a Bluetooth connection. In some examples, wireless communication links 163a-e may be a Wi-Fi (e.g., direct Wi-Fi) connection. [0071] The media adapter 107 may be any type of computing device that includes one or more processors (processor(s) 170), one or more memory devices (memory device(s) 172), and an operating system 174. In some implementations, the processor(s) 170 may include a system on a chip (SoC). The SoC may include a central processing unit (CPU), a graphic processing unit (GPU), one or more memory interfaces, and one or more input/output interfaces and devices. In some implementations, the operating system 174 may be system software that manages computer hardware, software resources, and provides common services for computing programs. [0072] The network-connected display device 104 may include the unified television application 130. The unified television application 130 may keep a record of the interactions of the user with the media content received from the server computer 106. The network- connected display device 104 may send the record of the interactions to the server computer 106 for use in determining media content recommendations for the user. [0073] In some implementations, the network-connected display device 104 may be configured to execute the unified television application 130. For example, the network- connected display device 104 may be a smart television. For example, a smart television may be a network-connected television that may connect to media content providers (e.g., media content providers 160) by way of a network (e.g., the network 150). The media content providers may source media content to the smart television. In these implementations, a user may interact with the unified television application 130 to access media content from the media content providers 160. The unified television application 130 may interface with the server computer 106, and specifically with the server-side TV application 116. The unified television application 130 may provide similar functionality to the user as that provided by an application executing on the media adapter 107. For example, executing the unified television application 130 by the network-connected display device 104 allows the network- Atty Docket No.0120-996WO1 connected display device 104 to obtain a media content recommendation stream from the server computer 106. [0074] The network-connected display device 104 may be configured to connect to the network 150. In some implementations, the network-connected display device 104 is a television (e.g., a smart television (TV)). The network-connected display device 104 may include one or more processors (processor(s) 156), one or more memory devices (memory device(s) 152), and an operating system (OS) 154. The operating system 154 may execute (or assist with executing) the unified television application 130. [0075] In some implementations, the operating system 154 may be a browser application. A browser application is a web browser configured to access information on the Internet by way of a network (e.g., the network 150). A browser application may launch one or more browser tabs in the context of one or more browser windows in the browser application. In some implementations, the operating system 154 is a Linux-based operating system configured to execute (or assist with executing) the unified television application 130. [0076] The system 100 may include one or more server computers (e.g., the server computer 106) configured to interface with the mobile computing device 102, the media adapter 107, the media content providers 160, and the network-connected display device 104 by way of the network 150. In some implementations, the network 150 may establish a wireless communication link between the network-connected display device 104, the mobile computing device 102, the media adapter 107, the media content providers 160, and the server computer 106. [0077] The server computer 106 may include the unified media platform (UMP) 158. The UMP 158 may facilitate the providing of media content items to the network-connected display device 104 as described herein. The server computer 106 may include the server- side TV application 116. The server-side TV application 116 may facilitate providing the media content items for playing on the network-connected display device 104. [0078] The mobile computing device 102 may include the mobile computing device display 108. In some implementations, the mobile computing device display 108 is a display device such as a liquid crystal display (LCD), a light-emitting diode (LED) display, an organic light-emitting diode (OLED) display, or an active-matrix organic light-emitting Atty Docket No.0120-996WO1 diode (AMOLED) display. The network-connected display device 104 may include the display 132. In some implementations, the display 132 is a display device such as a liquid crystal display (LCD), a light-emitting diode display (LED) display, a plasma display, a quantum dot light-emitting diode display (QLED) display, or an organic light-emitting diode (OLED) display. [0079] The processor(s) 156, the processor(s) 140, the processor(s) 170, and the processor(s) 180 may be formed in a substrate configured to execute one or more machine executable instructions or pieces of software, firmware, or a combination thereof. The processor(s) 156, the processor(s) 140, the processor(s) 170, and the processor(s) 180 may be semiconductor-based. For example, the processor(s) 156, the processor(s) 140, the processor(s) 170, and the processor(s) 180 may include semiconductor material that can perform digital logic. [0080] The memory device(s) 152, the memory device(s) 142, the memory device(s) 172, and the memory device(s) 182 may include main memory that stores information in a format that can be read and/or executed by the processor(s) 156, the processor(s) 140, the processor(s) 170, and the processor(s) 180 respectively. The memory device(s) 152, the memory device(s) 142, the memory device(s) 172, and the memory device(s) 182 may include one or more random-access memory (RAM) devices and/or one or more read-only memory (ROM) devices. [0081] The memory device(s) 152, memory device(s) 142, the memory device(s) 172, and the memory device(s) 182 may store applications that, when executed by the processor(s) 156, the processor(s) 140, the processor(s) 170, and the processor(s) 180, respectively, perform operations. For example, the memory device(s) 142 may store the operating system 144 and the TV application 110 that, when executed by the processor(s) 140, may perform operations on the mobile computing device 102. For example, the memory device(s) 152 may store the operating system 154 and the unified television application 130 that, when executed by the processor(s) 156, may perform operations on the network- connected display device 104. [0082] In some implementations, the memory device(s) 182 may represent any kind of (or multiple kinds of) memory (e.g., RAM, flash, cache, disk, tape, etc.). In some implementations, the memory device(s) 182 may include external storage, e.g., memory Atty Docket No.0120-996WO1 physically remote from but accessible by the server computer 106. The server computer 106 may include one or more modules, engines, or applications representing specially programmed software. In some implementations, the server computer 106 may include the operating system 184, the server-side TV application 116, the knowledge module 166, the AI module 194 the includes the Gen AI engine 162 and the generative AI model(s) 164, the UMP 158, the images repository 168, the image-to-image synthesizer 176, processor(s) 180, and the memory device(s) 182. For example, the memory device(s) 182 may store the operating system 184, the server-side TV application 116, the knowledge module 166, the AI module 194 including the generative AI engine 162 and the generative AI model(s) 164, the UMP 158, the images repository 168, and the image-to-image synthesizer 176 that, when executed by the processor(s) 180, may perform operations on server computer 106 to implement one or more of the methods and processes described herein. [0083] The network 150 may include the Internet and/or other types of data networks, such as a local area network (LAN), a wide area network (WAN), a cellular network, satellite network, or other types of data networks. The network 150 may also include any number of computing devices (e.g., computer, servers, routers, network switches, etc.) that are configured to receive and/or transmit data within the network 150. The network 150 may further include any number of hardwired and/or wireless connections. The network 150 may be, for example, communications networks having one or more types of topologies, including but not limited to the Internet, intranets, local area networks (LANs), cellular networks, Ethernet, Storage Area Networks (SANs), telephone networks, and Bluetooth personal area networks (PAN). In some implementations, two or more devices in a sub-network may be coupled by way of a wired connection, while at least some of the devices in the same sub-network are coupled by way of a local radio communication network (e.g., ZigBee, Z-Wave, Insteon, Bluetooth, Wi-Fi and other radio communication networks). [0084] FIG. 2 is an illustration of an example process 200 for generating a customized artwork (e.g., output image 204) based on an input image 202. The input image 202 can be selected from a carousel of features images (e.g., the featured images carousel 134). Referring to FIGS. 1A-C, for example, the image-to-image synthesizer 176 may perform the process 200. The process 200 includes the image-to-image synthesizer 176 using the ensemble based generative AI model 178 to create or generate the output image 204. The Atty Docket No.0120-996WO1 process 200 may synthesize or collect a dataset for generating the output image 204 based on the input image 202. The image-to-image synthesizer 176 can use the ensemble based generative AI model 178 to extract information and data related to the content of the input image 202. For example, the ensemble based generative AI model 178 may include an image comprehension model 208. The image comprehension model 208 may receive the input image 202 as input and determine an image comprehension keyword token 210. The image comprehension keyword token is a keyword that identifies a key part, component, or object in the input image 202. For example, if the input image 202 is an image of a frog, the image comprehension keyword token 210 may be “frog”. In another example, if the input image 202 includes snowmen, the image comprehension keyword token 210 may be “snowmen”. In another example, if the input image 202 is of a cat in a spacesuit, the image comprehension keyword token 210 may be “cat”. [0085] The image comprehension model 208 may be a family of large language models (LLMs) that can recognize and interpret the content of the input image 202. The image comprehension model 208 may determine at least one image comprehension keyword token 210 associated with the input image 202. In addition, or in the alternative, the image comprehension model 208 may determine additional keyword tokens 206a-e associated with additional information and data for the input image 202. [0086] The ensemble based generative AI model 178 may combine the image comprehension keyword token 210 and the keyword tokens 206a-e to form a homologated keyword token 212. The homologated keyword token 212 may be representative of the content of the input image 202. The homologated keyword token 212 may be considered an official keywork token for associating with the input image 202. [0087] FIG.3A is an illustration of a first image 300 that includes three snowmen. FIG. 3B is an illustration of a second image 350 that is based on the first image 300. For example, referring to FIGS. 1A-C and 2, the first image 300 may be on a sweater or scarf that belongs to a user (e.g., the user 101). The first image 300 may be included in the images repository 168. For example, the input image 202 may be the first image 300. The input image 202 (the first image 300) may be input to the image comprehension model 208 of the ensemble based generative AI model 178. The image comprehension model 208 may output the image comprehension keyword token 210 (e.g., “snowmen”) in addition to the keyword Atty Docket No.0120-996WO1 tokens 206a-e (e.g., “three”, “wearing”, “hats”, “and”, “scarves”, respectively). The image comprehension keyword token 210 and the keyword tokens 206a-e are formed into a homologated keyword token 212. The image-to-image synthesizer 176 generates the homologated keyword token 212 for use as a basis for generating the output image 204. [0088] The homologated keyword token 212 may be input to an image generation model 214. The image generation model 214 may use a text-to-image diffusion model that incorporates a degree of photorealism and a deep level of language understanding to generate the output image 204. For example, the user 101 may request an artistic rendition of the input image 202 (e.g., the first image 300 as shown in FIG.3A). As described herein, the artistic rendition may be based on an artistic style preferred by the user. The artistic style may be determined from and/or based on an artwork or the works of an artist that the user likes or that the user has selected for use by the image-to-image synthesizer 176 as an artistic basis for generating the output image 204. [0089] The image-to-image synthesizer 176 using one or more of the processes, methods, and systems described herein may generate the output image 204 as image 350 as shown in FIG.3B. FIG.3C is an illustration of an ambient screen (e.g., the ambient screen 109) of a smart TV (e.g., the network-connected display device 104) displaying the output image 204 as a customized artwork 313. For example, the ambient screen 109 may show the output image 204 along with the entries 111a-d. The output image 204 may be displayed on the ambient screen 109 of the network-connected display device 104. The output image 204 may be considered a customized artwork. [0090] FIG. 4 is an illustration of another example process 400 for generating an artwork (e.g., output image 404). The process 400 may be for an image-to-image synthesizer 476 that uses keyword tokens to synthesize or collect a dataset for generating the output image 404 based on inputs to a generative AI (Gen AI) model 406. For example, referring to FIG.1B, the Gen AI model 406 may be included in the Gen AI model(s) 164. [0091] In some implementations, the Gen AI model 406 may receive images (e.g., first image 420a and second image 420b) as input images from a web crawler 424. The web crawler 424 may provide the images 420a-b that include the same or a similar object (e.g., two images that each include a frog). The Gen AI model 406 may include an image comprehension model 408. The image comprehension model 408 may receive the images Atty Docket No.0120-996WO1 420a-b as input and determine an image comprehension keyword token 410 common to both images 420a-b. The image comprehension model 208 may be a family of large language models (LLMs) that can recognize and interpret the content of the images 420a-b. The image comprehension model 208 may determine at least one image comprehension keyword token 410 associated with each of the images 420a-b. In addition, or in the alternative, the image comprehension model 208 may determine additional keyword tokens 416a-b associated with additional information and data for the first image 420a and additional keyword tokens 418a- b associated with additional information and data for the second image 420b. [0092] For example, the Gen AI model 406 may receive the images 420a-b and determine that there is a common object included in each image (e.g., a frog). The image comprehension model 408 may output the image comprehension keyword token 410 (e.g., “frog”) that identifies a common object between the images 420a-b (e.g., a frog). The image comprehension keyword token 410 for the identified common object between the images 420a-b may be used as a keyword token for each image (e.g., first image keyword token 416a (e.g., “frog”) for the first image 420a, and second image keyword token 418a (e.g., “frog”) for the second image 420b). The image comprehension model 408 may provide additional keyword tokens for each image (e.g., first image additional keyword token(s) 416b for the first image 420a and second image additional keyword token(s) 418b for the second image 420b) that are differentiators between first image 420a and second image 420b. [0093] The Gen AI model 406 may determine the differentiating keyword tokens, for example, by analyzing two images that include a similar or the same object (e.g., a frog) retrieved by performing a web crawl of available images for the object (e.g., images of frogs available on the web). For example, the Gen AI model 406 may include a keyword-based summarizer 412 that determines keyword-based summarizations for two images that include the same object (e.g., first image 420a and second image 420b where the object is a frog). The keyword-based summarizer 412 may generate keyword-based summarizations for each image. For example, where the common object between the first image 420a and the second image 420b is a frog, the keyword-based summarizer 412 may generate the keyword-based summaries for each image as shown in Equation 1 for the first image 420a (Sum[Frog1_Image]) and Equation 2 for the second image 420b (Sum[Frog2_Image]). Atty Docket No.0120-996WO1 [0094] Equation 1: Sum[Img(Frog1)] = [keyword[frog] + [Keyword 11] … [Keyword 1n]] where keyword[frog] = the first image keyword token 416a, [Keyword 1] ... [Keyword n] are the first image additional keyword token(s) 416b. [0095] Equation 2: Sum[Img(Frog2)] = [keyword[frog] + [Keyword 21]... [Keyword 2n]] where keyword[frog] = the second image keyword token 418a, [Keyword 21] ... [Keyword 2n] are the second image additional keyword token(s) 418b. [0096] In some implementations, the Gen AI model 406 may include an image generation model 414. The Gen AI model 406 may use the image generation model 414 to generate an image of the common object as included in the second image 420b (e.g., the frog in the second image 420b) based on the differentiating keyword tokens for the first image 420a (e.g., first image additional keyword token(s) 416b) as the output image 404. Equation 3 is an example equation showing the keyword tokens for use in generating the output image 404 that includes the common object (e.g., the frog). [0097] Equation 3: Keyword Tokens = [Keyword 21] + (neg)[Keyword 11] + ... [Keyword 1n]] [0098] The process 200 and the process 400 may provide datasets for output images that have a fixed dimensional output (e.g., a fixed aspect ratio). For example, the dimensions of images for display on the screen of a smart TV are typically in a 16:9 aspect ratio. In addition, or in the alternative, process 200 and process 400 may also provide additional words and/or images that fit into the fixed dimensional output. [0099] In some implementations, large language models (LLMs) may utilize models that include billions of parameters. In addition, or in the alternative, the LLMs may have a latency time that may be limiting for an end user of the system utilizing the model. Style transfer based generative adversarial networks (GANs) and cyclic GANs may be suited for use by image-to-image synthesizers. For example, referring to FIGS. 2 and 4, the style- transfer based GANs may be used by an image generation model (e.g., the image generation model 214, the image generation model 414) included in an image-to-image synthesizer (e.g., the image-to-image synthesizer 176, the image-to-image synthesizer 476). The image generation model (e.g., the image generation model 214, the image generation model 414) Atty Docket No.0120-996WO1 may be finetuned and refined using intelligent-pruning and reinforcement learning (RL) based model compression techniques. The refining and finetuning of an image generation model may reduce the number of parameters typically used by LLMs (on the order of billions) by approximately 99%. In these implementations, this significant reduction in the number of parameters used by the LLMs can allow the image generation model to run on a smart TV or a low-cost backend server. Therefore, in some implementations, the image-to- image synthesizer (e.g., the image-to-image synthesizer 176, the image-to-image synthesizer 476) can operate on a network-connected display device (e.g., the network-connected display device 104) such as a smart TV. [00100] In some implementations, a user may provide information and data that may be used for preemptive intelligent suggestions for keyword tokens. For example, a user may verbally articulate preferences for output images using, for example, a voice assistant on the smart TV. In another example, preferences for output images may be based on past user experience, watch history, regional trends, etc. In some implementations, referring to FIG. 1B, the network-connected display device 104 may interface with the knowledge module 166 on the server computer 106 to obtain user preferences. [0061] FIG.5 illustrates a flowchart depicting example operations of generating an image (e.g., a customized artwork) for display in an ambient screen of a television application executing on a display device according to implementations described throughout this disclosure. Although the flowchart 500 of FIG.5 illustrates the operations in sequential order, it will be appreciated that this is merely an example, and that additional or alternative operations may be included. Further, operations of FIG. 5 and related operations may be executed in a different order than that shown, or in a parallel or overlapping fashion. The operations may define a computer-implemented method. Although the flowchart 500 is described with reference to the system 100 of FIG. 1B, the flowchart 500 may be executed according to any of the figures discussed herein. In some examples, the operations of the flowchart 500 are executed by server computer 106. [0062] Operation 510 includes sending, by a server computer to a display device, images for including in a carousel on the display device, the images included in the carousel for use as a basis for generating customized artwork for display in an ambient screen of a television application executing on the display device. For example, the server computer 106 Atty Docket No.0120-996WO1 may send images from the images repository 168 to the network-connected display device 104 for including in the featured images carousel 134 for use as a basis for generating customized artwork (e.g., customized artwork 113) for display in the ambient screen 109 of the unified television application 130. [0063] Operation 520 includes generating an output image including receiving a selection of an input image from the carousel, extracting image content information and data from the input image, determining keyword tokens for associating with the input image; and applying an ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens. [0064] Operation 530 includes sending, by the server computer to the display device, the output image as the customized artwork for display in the ambient screen of the television application executing on the display device. For example, the image-to-image synthesizer 176 may apply the ensemble based generative AI model 178 to an input image (e.g., input image 202) to generate the output image 204. The server computer 106 may send the output image 204 to the network-connected display device 104 for display in the ambient screen 109 of the unified television application 130. [0065] The following examples can be combined with one another in any suitable combination. Features and examples described herein with respect to the method can be implemented in non-transitory computer-readable medium and/or by the system, and vice versa. [0066] In some examples, the techniques described herein relate to a method including: sending, by a server computer to a display device, images for including in a carousel on the display device, the images included in the carousel for use as a basis for generating customized artwork for display in an ambient screen of a television application executing on the display device; generating an output image including: receiving a selection of an input image from the carousel; extracting image content information and data from the input image; determining keyword tokens for associating with the input image; and applying an ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens; and sending, by the server computer to the display device, the output image as the customized artwork for display in the ambient screen of the television application executing on the display device. Atty Docket No.0120-996WO1 [0067] In some examples, the techniques described herein relate to a method, wherein extracting image content information and data from the input image includes extracting the image content information and data from the input image using generative artificial intelligence. [0068] In some examples, the techniques described herein relate to a method, wherein generating the output image further includes applying a generative artificial intelligence model that includes an image comprehension model and an image generation model. [0069] In some examples, the techniques described herein relate to a method, wherein the ensemble of style transfer networks and generative adversarial networks are included in the image generation model. [0070] In some examples, the techniques described herein relate to a method, wherein extracting image content information and data from the input image includes extracting the image content information and data from the input image by the image comprehension model. [0071] In some examples, the techniques described herein relate to a method, wherein the keyword tokens include an image comprehension keyword token; and wherein the image comprehension model determines the image comprehension keyword token for the input image based on the image content information and data for the input image. [0072] In some examples, the techniques described herein relate to a method, wherein the image generation model generates the output image as an artistic rendering of the input image. [0073] In some examples, the techniques described herein relate to a method, wherein the keyword tokens include an image comprehension keyword token and at least one additional keyword token; and wherein generating the output image further includes generating a homologated keyword token based on the image comprehension keyword token and the at least one additional keyword token. [0074] In some examples, the techniques described herein relate to a method, wherein applying the ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens includes Atty Docket No.0120-996WO1 applying the ensemble of style transfer networks and generative adversarial networks on the homologated keyword token. [0075] In some examples, the techniques described herein relate to a method, wherein the output image is generated by an image-to-image synthesizer. [0076] In some examples, the techniques described herein relate to a non-transitory computer-readable medium storing executable instructions that when executed by at least one processor of a server computer cause the at least one processor to execute operations, the operations including: sending, to a display device, images for including in a carousel on the display device, the images included in the carousel for use as a basis for generating customized artwork for display in an ambient screen of a television application executing on the display device; generating an output image including: receiving a selection of an input image from the carousel; extracting image content information and data from the input image; determining keyword tokens for associating with the input image; and applying an ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens; and sending, to the display device, the output image as the customized artwork for display in the ambient screen of the television application executing on the display device. [0077] In some examples, the techniques described herein relate to a non-transitory computer-readable medium, wherein extracting image content information and data from the input image includes extracting the image content information and data from the input image using generative artificial intelligence. [0078] In some examples, the techniques described herein relate to a non-transitory computer-readable medium, wherein generating the output image further includes applying a generative artificial intelligence model that includes an image comprehension model and an image generation model. [0079] In some examples, the techniques described herein relate to a non-transitory computer-readable medium, wherein the ensemble of style transfer networks and generative adversarial networks are included in the image generation model. [0080] In some examples, the techniques described herein relate to a non-transitory computer-readable medium, wherein extracting image content information and data from the Atty Docket No.0120-996WO1 input image includes extracting the image content information and data from the input image by the image comprehension model. [0081] In some examples, the techniques described herein relate to a non-transitory computer-readable medium, wherein the keyword tokens include an image comprehension keyword token; and wherein the image comprehension model determines the image comprehension keyword token for the input image based on the image content information and data for the input image. [0082] In some examples, the techniques described herein relate to a non-transitory computer-readable medium, wherein the image generation model generates the output image as an artistic rendering of the input image. [0083] In some examples, the techniques described herein relate to a non-transitory computer-readable medium, wherein the keyword tokens include an image comprehension keyword token and at least one additional keyword token; and wherein generating the output image further includes generating a homologated keyword token based on the image comprehension keyword token and the at least one additional keyword token. [0084] In some examples, the techniques described herein relate to a non-transitory computer-readable medium, wherein applying the ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens includes applying the ensemble of style transfer networks and generative adversarial networks on the homologated keyword token. [0085] In some examples, the techniques described herein relate to a non-transitory computer-readable medium, wherein the output image is generated by an image-to-image synthesizer. [0086] In some examples, the techniques described herein relate to a system including: at least one processor; and a non-transitory computer-readable medium storing instructions that when executed by the at least one processor cause the system to: send, from a server computer to a display device, images for including in a carousel on the display device, the images included in the carousel for use as a basis for generating customized artwork for display in an ambient screen of a television application executing on the display device; generate an output image including: receiving a selection of an input image from the carousel; extracting image content information and data from the input image; determining Atty Docket No.0120-996WO1 keyword tokens for associating with the input image; and applying an ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens; and send, from the server computer to the display device, the output image as the customized artwork for display in the ambient screen of the television application executing on the display device. [0087] In some examples, the techniques described herein relate to a system, wherein extracting image content information and data from the input image includes extracting the image content information and data from the input image using generative artificial intelligence. [0088] In some examples, the techniques described herein relate to a system, wherein generating the output image further includes applying a generative artificial intelligence model that includes an image comprehension model and an image generation model. [0089] In some examples, the techniques described herein relate to a system, wherein the ensemble of style transfer networks and generative adversarial networks are included in the image generation model. [0090] In some examples, the techniques described herein relate to a system, wherein extracting image content information and data from the input image includes extracting the image content information and data from the input image by the image comprehension model. [0091] In some examples, the techniques described herein relate to a system, wherein the keyword tokens include an image comprehension keyword token; wherein the image comprehension model determines the image comprehension keyword token for the input image based on the image content information and data for the input image. [0092] In some examples, the techniques described herein relate to a system, wherein the image generation model generates the output image as an artistic rendering of the input image. [0093] In some examples, the techniques described herein relate to a system, wherein the keyword tokens include an image comprehension keyword token and at least one additional keyword token; and wherein generating the output image further includes generating a homologated keyword token based on the image comprehension keyword token and the at least one additional keyword token. Atty Docket No.0120-996WO1 [0094] In some examples, the techniques described herein relate to a system, wherein applying the ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens includes applying the ensemble of style transfer networks and generative adversarial networks on the homologated keyword token. [0095] In some examples, the techniques described herein relate to a system, wherein the output image is generated by an image-to-image synthesizer. [0096] Various implementations of the systems and techniques described here can be realized in digital electronic circuitry, integrated circuitry, specially designed ASICs (application specific integrated circuits), computer hardware, firmware, software, and/or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device. [0097] These computer programs (also known as programs, software, software applications or code) include machine instructions for a programmable processor and can be implemented in a high-level procedural and/or object-oriented programming language, and/or in assembly/machine language. As used herein, the terms “machine-readable medium” “computer-readable medium” refers to any computer program product, apparatus and/or device (e.g., magnetic discs, optical disks, memory, Programmable Logic Devices (PLDs)) used to provide machine instructions and/or data to a programmable processor, including a non-transitory machine-readable medium that receives machine instructions as a machine-readable signal. The term “machine-readable signal” refers to any signal used to provide machine instructions and/or data to a programmable processor. A computer program product comprising computer-executable instructions which, when executed by at least one computing apparatus, cause the at least one computing apparatus to perform the method described herein may also be provided. [0098] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a Atty Docket No.0120-996WO1 keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input. [0099] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a client computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or non-transitory medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (“LAN”), a wide area network (“WAN”), and the Internet. [00100] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. [00101] In this specification and the appended claims, the singular forms "a," "an" and "the" do not exclude the plural reference unless the context clearly dictates otherwise. Further, conjunctions such as “and,” “or,” and “and/or” are inclusive unless the context clearly dictates otherwise. For example, “A and/or B” includes A alone, B alone, and A with B. Further, connecting lines or connectors shown in the various figures presented are intended to represent example functional relationships and/or physical or logical couplings between the various elements. Many alternative or additional functional relationships, physical connections or logical connections may be present in a practical device. Moreover, no item or component is essential to the practice of the embodiments disclosed herein unless the element is specifically described as “essential” or “critical”. [00102] Terms such as, but not limited to, approximately, substantially, generally, etc. are used herein to indicate that a precise value or range thereof is not required and need not Atty Docket No.0120-996WO1 be specified. As used herein, the terms discussed above will have ready and instant meaning to one of ordinary skill in the art. [00103] Moreover, use of terms such as up, down, top, bottom, side, end, front, back, etc. herein are used with reference to a currently considered or illustrated orientation. If they are considered with respect to another orientation, it should be understood that such terms must be correspondingly modified. [00104] Further, in this specification and the appended claims, the singular forms "a," "an" and "the" do not exclude the plural reference unless the context clearly dictates otherwise. Moreover, conjunctions such as “and,” “or,” and “and/or” are inclusive unless the context clearly dictates otherwise. For example, “A and/or B” includes A alone, B alone, and A with B. [00105] Although certain example methods, apparatuses and articles of manufacture have been described herein, the scope of coverage of this patent is not limited thereto. It is to be understood that terminology employed herein is for the purpose of describing particular aspects and is not intended to be limiting. On the contrary, this patent covers all methods, apparatus and articles of manufacture fairly falling within the scope of the claims of this patent. [00106] Further to the descriptions above, a user may be provided with controls allowing the user to make an election as to both if and when systems, programs, or features described herein may enable collection of user information (e.g., a user’s preferences, a user’s current location, a user’s credentials, etc.), and if the user is sent content or communications from a server. In addition, certain data may be treated in one or more ways before it is stored or used, so that personally identifiable information is removed. For example, a user’s identity may be treated so that no personally identifiable information can be determined for the user, or a user’s geographic location may be generalized where location information is obtained (such as to a city, ZIP code, or state level), so that a particular location of a user cannot be determined. Thus, the user may have control over what information is collected about the user, how that information is used, and what information is provided to the user.

Claims

Atty Docket No.0120-996WO1 WHAT IS CLAIMED IS: 1. A method comprising: sending, by a server computer to a display device, images for including in a carousel on the display device, the images included in the carousel for use as a basis for generating customized artwork for display in an ambient screen of a television application executing on the display device; generating an output image comprising: receiving a selection of an input image from the carousel; extracting image content information and data from the input image; determining keyword tokens for associating with the input image; and applying an ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens; and sending, by the server computer to the display device, the output image as the customized artwork for display in the ambient screen of the television application executing on the display device. 2. The method of claim 1, wherein extracting image content information and data from the input image comprises extracting the image content information and data from the input image using generative artificial intelligence. 3. The method of claim 1 or claim 2, wherein generating the output image further comprises applying a generative artificial intelligence model that includes an image comprehension model and an image generation model. 4. The method of claim 3, wherein the ensemble of style transfer networks and generative adversarial networks are included in the image generation model. 5. The method of claim 3 or claim 4, wherein extracting image content information and data from the input image comprises extracting the image content information and data from the input image by the image comprehension model. Atty Docket No.0120-996WO1 6. The method of any of claims 3 to 5, wherein the keyword tokens include an image comprehension keyword token; and wherein the image comprehension model determines the image comprehension keyword token for the input image based on the image content information and data for the input image. 7. The method of any of claims 3 to 6, wherein the image generation model generates the output image as an artistic rendering of the input image. 8. The method of claim 1, wherein the keyword tokens include an image comprehension keyword token and at least one additional keyword token; and wherein generating the output image further comprises generating a homologated keyword token based on the image comprehension keyword token and the at least one additional keyword token. 9. The method of claim 8, wherein applying the ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens comprises applying the ensemble of style transfer networks and generative adversarial networks on the homologated keyword token. 10. The method of any of claims 1 to 9, wherein the output image is generated by an image-to-image synthesizer. 11. A non-transitory computer-readable medium storing executable instructions that when executed by at least one processor of a server computer cause the at least one processor to execute operations, the operations comprising: sending, to a display device, images for including in a carousel on the display device, the images included in the carousel for use as a basis for generating customized artwork for display in an ambient screen of a television application executing on the display device; Atty Docket No.0120-996WO1 generating an output image comprising: receiving a selection of an input image from the carousel; extracting image content information and data from the input image; determining keyword tokens for associating with the input image; and applying an ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens; and sending, to the display device, the output image as the customized artwork for display in the ambient screen of the television application executing on the display device. 12. The non-transitory computer-readable medium of claim 11, wherein extracting image content information and data from the input image comprises extracting the image content information and data from the input image using generative artificial intelligence. 13. The non-transitory computer-readable medium of claim 11 or claim 12, wherein generating the output image further comprises applying a generative artificial intelligence model that includes an image comprehension model and an image generation model. 14. The non-transitory computer-readable medium of claim 13, wherein the ensemble of style transfer networks and generative adversarial networks are included in the image generation model. 15. The non-transitory computer-readable medium of claim 13 or claim 14, wherein extracting image content information and data from the input image comprises extracting the image content information and data from the input image by the image comprehension model. 16. The non-transitory computer-readable medium of any of claims 13 to 15, wherein the keyword tokens include an image comprehension keyword token; and wherein the image comprehension model determines the image comprehension keyword token for the input image based on the image content information and data for the input image. Atty Docket No.0120-996WO1 17. The non-transitory computer-readable medium of any of claims 13 to 16, wherein the image generation model generates the output image as an artistic rendering of the input image. 18. The non-transitory computer-readable medium of claim 11, wherein the keyword tokens include an image comprehension keyword token and at least one additional keyword token; and wherein generating the output image further comprises generating a homologated keyword token based on the image comprehension keyword token and the at least one additional keyword token. 19. The non-transitory computer-readable medium of claim 18, wherein applying the ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens comprises applying the ensemble of style transfer networks and generative adversarial networks on the homologated keyword token. 20. The non-transitory computer-readable medium of any of claims 11 to 19, wherein the output image is generated by an image-to-image synthesizer. 21. A system comprising: at least one processor; and a non-transitory computer-readable medium storing instructions that when executed by the at least one processor cause the system to: send, from a server computer to a display device, images for including in a carousel on the display device, the images included in the carousel for use as a basis for generating customized artwork for display in an ambient screen of a television application executing on the display device; generate an output image comprising: receiving a selection of an input image from the carousel; Atty Docket No.0120-996WO1 extracting image content information and data from the input image; determining keyword tokens for associating with the input image; and applying an ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens; and send, from the server computer to the display device, the output image as the customized artwork for display in the ambient screen of the television application executing on the display device. 22. The system of claim 21, wherein extracting image content information and data from the input image comprises extracting the image content information and data from the input image using generative artificial intelligence. 23. The system of claim 21 or claim 22, wherein generating the output image further comprises applying a generative artificial intelligence model that includes an image comprehension model and an image generation model. 24. The system of claim 23, wherein the ensemble of style transfer networks and generative adversarial networks are included in the image generation model. 25. The system of claim 23 or claim 24, wherein extracting image content information and data from the input image comprises extracting the image content information and data from the input image by the image comprehension model. 26. The system of any of claims 23 to 25, wherein the keyword tokens include an image comprehension keyword token; wherein the image comprehension model determines the image comprehension keyword token for the input image based on the image content information and data for the input image. 27. The system of any of claims 23 to 26, wherein the image generation model generates the output image as an artistic rendering of the input image. Atty Docket No.0120-996WO1 28. The system of claim 21, wherein the keyword tokens include an image comprehension keyword token and at least one additional keyword token; and wherein generating the output image further comprises generating a homologated keyword token based on the image comprehension keyword token and the at least one additional keyword token. 29. The system of claim 28, wherein applying the ensemble of style transfer networks and generative adversarial networks on the image content information and data using the keyword tokens comprises applying the ensemble of style transfer networks and generative adversarial networks on the homologated keyword token. 30. The system of any of claims 21 to 29, wherein the output image is generated by an image-to-image synthesizer.
PCT/US2024/033840 2024-06-13 2024-06-13 Content for displaying in an ambient mode screen for a television application Pending WO2025259282A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/US2024/033840 WO2025259282A1 (en) 2024-06-13 2024-06-13 Content for displaying in an ambient mode screen for a television application

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/US2024/033840 WO2025259282A1 (en) 2024-06-13 2024-06-13 Content for displaying in an ambient mode screen for a television application

Publications (1)

Publication Number Publication Date
WO2025259282A1 true WO2025259282A1 (en) 2025-12-18

Family

ID=91898217

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2024/033840 Pending WO2025259282A1 (en) 2024-06-13 2024-06-13 Content for displaying in an ambient mode screen for a television application

Country Status (1)

Country Link
WO (1) WO2025259282A1 (en)

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022075533A1 (en) * 2020-10-07 2022-04-14 Samsung Electronics Co., Ltd. Method of on-device generation and supplying wallpaper stream and computing device implementing the same

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022075533A1 (en) * 2020-10-07 2022-04-14 Samsung Electronics Co., Ltd. Method of on-device generation and supplying wallpaper stream and computing device implementing the same

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
BENJAMIN Z YAO ET AL: "I2T: Image Parsing to Text Description", PROCEEDINGS OF THE IEEE, IEEE. NEW YORK, US, vol. 98, no. 8, 17 June 2010 (2010-06-17), pages 1485 - 1508, XP011311730, ISSN: 0018-9219 *
LÓPEZ-SÁNCHEZ MARCO ET AL: "Supervised Deep Learning Techniques for Image Description: A Systematic Review", ENTROPY, vol. 25, no. 4, 23 March 2023 (2023-03-23), CH, pages 553, XP093228700, ISSN: 1099-4300, DOI: 10.3390/e25040553 *
OZGEN AZMI CAN ET AL: "Text-to-Painting on a Large Variance Dataset with Sequential Generative Adversarial Networks", 2020 28TH SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS CONFERENCE (SIU), IEEE, 5 October 2020 (2020-10-05), pages 1 - 4, XP033872811, DOI: 10.1109/SIU49456.2020.9302112 *

Similar Documents

Publication Publication Date Title
CN113746874B (en) A voice package recommendation method, device, equipment and storage medium
US11899907B2 (en) Method, apparatus and device for displaying followed user information, and storage medium
US10922327B2 (en) Search guidance
US8245124B1 (en) Content modification and metadata
US10216841B2 (en) System for referring to and/or embedding posts, videos or digital media within other posts, videos or digital media and posts within any part of another posts, videos or digital media
WO2021052085A1 (en) Video recommendation method and apparatus, electronic device and computer-readable medium
US20100312596A1 (en) Ecosystem for smart content tagging and interaction
JP5623516B2 (en) Open search support
US20240292052A1 (en) Page display method and apparatus, electronic device and storage medium
US20130047123A1 (en) Method for presenting user-defined menu of digital content choices, organized as ring of icons surrounding preview pane
US20170220591A1 (en) Modular search object framework
US20150317354A1 (en) Intent based search results associated with a modular search object framework
KR20130014581A (en) Selecting content based on interest tags that are included in an interest cloud
WO2021103398A1 (en) Smart television and server
US20180174110A1 (en) Personalized Calendar for Digital Media Content-Related Events
US20150317319A1 (en) Enhanced search results associated with a modular search object framework
JP2012003759A (en) System and method enabling visual filtering of content
CN105760521A (en) Information input method and device
US20090222276A1 (en) Apparatus, System, and Method for Cloning Web Page Designs or Avatar Designs
CN106471819A (en) System and method for improving the accuracy in media asset recommended models
CN107105030A (en) Promotional content method for pushing and device
CN113228003B (en) Serving different content pages based on varying user interactions with a single content item
WO2021243985A1 (en) Method and apparatus for generating weather forecast video, electronic device, and storage medium
US20140351000A1 (en) Dynamic Modification of A Parameter of An Image Based on User Interest
WO2023035893A9 (en) Search processing method and apparatus, and device, medium and program product

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 24740681

Country of ref document: EP

Kind code of ref document: A1