WO2021169134A1 - 表情包生成方法、装置、设备和介质 - Google Patents

表情包生成方法、装置、设备和介质 Download PDF

Info

Publication number
WO2021169134A1
WO2021169134A1 PCT/CN2020/100034 CN2020100034W WO2021169134A1 WO 2021169134 A1 WO2021169134 A1 WO 2021169134A1 CN 2020100034 W CN2020100034 W CN 2020100034W WO 2021169134 A1 WO2021169134 A1 WO 2021169134A1
Authority
WO
WIPO (PCT)
Prior art keywords
emoticon
text
information
package
matching
Prior art date
Application number
PCT/CN2020/100034
Other languages
English (en)
French (fr)
Inventor
徐相龙
朱剑锋
崔家华
向静
李红涛
韩琛
林书妃
苏莹
李世操
李慧琴
甘小楚
高菲
杨佳乐
麻雪云
李国洪
Original Assignee
北京百度网讯科技有限公司
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 北京百度网讯科技有限公司 filed Critical 北京百度网讯科技有限公司
Priority to JP2021516910A priority Critical patent/JP7212770B2/ja
Priority to KR1020217009336A priority patent/KR102598496B1/ko
Priority to US17/280,142 priority patent/US11521340B2/en
Priority to EP20864289.2A priority patent/EP3901786A4/en
Publication of WO2021169134A1 publication Critical patent/WO2021169134A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/51—Indexing; Data structures therefor; Storage structures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/334—Query execution
    • G06F16/3344—Query execution using natural language analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/335—Filtering based on additional data, e.g. user or group profiles
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/5866—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, manually generated location and time information
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/20—Movements or behaviour, e.g. gesture recognition

Definitions

  • the embodiments of the application relate to the field of image processing, such as Internet technology. Specifically, the embodiments of the present application provide a method, apparatus, device, and medium for generating an emoticon package.
  • the products that provide emoticon resources for Internet users are mainly emoticon search, that is, users can manually enter text, and the system will automatically match related emoticons for it.
  • the embodiments of the present application provide a method, device, equipment, and medium for generating an emoticon package to realize automatic and accurate generation of an emoticon package.
  • An embodiment of the present application provides a method for generating an emoticon package, which includes:
  • the target matching text is superimposed on the emoticon graph to generate a new emoticon package.
  • the related text of the emoticon and/or the similar emoticon package of the emoticon are determined, and the related text of the emoticon and/or the related text of the similar emoticon package are used as candidate matching texts to realize automatic determination of candidate matching texts.
  • the target matching text By determining the target matching text from the candidate matching text, because the associated text of the emoticon includes at least one of subject information, scene information, emotion information, action information, and connotation information, and the similar emoticon package and emoticon have similar characteristics, so The determined candidate matching text describes the content information of the emoticon. Therefore, compared to determining the target matching text from the preset text, the target matching text determined in the embodiment of the present application can describe the emoticon more accurately, that is, the accuracy of the target matching text is improved. Finally, by superimposing the accurately determined target matching text on the emoticon, the automatic and accurate generation of emoticons is realized.
  • determining the associated text of the emoticon includes:
  • the associated text of the emoticon is determined.
  • the embodiment of the present application determines the associated text of the emoticon based on the associated text of the similar emoticon package, thereby enriching the determined dimension of the associated text of the emoticon.
  • determining the associated text of the emoticon according to the associated text of the similar emoticon package includes:
  • the embodiment of the present application determines the target text from the associated text of similar emoticons based on the frequency of word usage; determines the associated text of the emoticon from the target text, thereby improving the accuracy of the associated text of the emoticon.
  • determining the associated text of the emoticon from the target text includes:
  • the repetitive text is filtered out from the recognition result of the emoticon and the target text to obtain the associated text of the emoticon, and the recognition result of the emoticon includes emotion information and subject information.
  • the embodiment of the present application integrates the recognition result of the emoticon and the target text, and uses the fused text as the associated text of the emoticon to enrich the text content of the associated text.
  • the associated text determined based on the recognition result of the emoticon includes at least two dimensions of information, that is, the text dimension of the associated text is enriched. Therefore, based on the determination method of the embodiment of the present application, the associated text of the determined emoticon is richer and diversified.
  • the similar emoticon package for determining the emoticon image includes:
  • the existing emoticon package is used as the similar emoticon package.
  • the embodiment of the present application determines similar emoticons from the two information dimensions of image information and associated text information, thereby improving the recall rate of similar emoticons, and then improving the evaluation of similar emoticons based on the associated text of similar emoticons.
  • the recall rate of the candidate matching text of the emoticon is improved.
  • the image information includes: emoticon package category information and object category information.
  • the embodiment of the present application determines similar emoticons from existing emoticons based on emoticon category information and object category information, thereby improving the accuracy of similar emoticons.
  • determining the target matching text from the associated text of the emoticon image and/or the associated text of the similar emoticon package includes:
  • the target matching text is determined from the candidate matching text, the candidate matching text including the associated text of the emoticon and/or the associated text of the similar emoticon package.
  • the embodiment of the present application determines target matching text from candidate matching texts based on at least one of word usage frequency, word length, and word semantics, thereby further improving the accuracy of determining target matching text.
  • the determining the target matching text from the candidate matching text based on the word length includes:
  • the embodiment of the present application determines whether the word length of the candidate matching text belongs to the set length range, and determines the target matching text from the candidate matching text, thereby achieving accurate determination of the target matching text.
  • determining the target matching text from the candidate matching text based on word semantics includes:
  • the word semantics of the candidate matching text includes word information of at least one of emotional words, hot words, and entity words;
  • the target matching text is determined from the candidate matching texts.
  • the embodiment of the present application determines the target matching text from the candidate matching text by determining the target matching text from the candidate matching text according to the semantic information of the word semantics of the candidate matching text including at least one of emotional words, hot words and entity words, thereby achieving the target matching text The exact determination.
  • the method further includes:
  • an emoticon is determined from the video image.
  • the embodiment of the present application recognizes the speech and/or action execution amplitude of the extracted video image; according to the recognition result, the expression image is determined from the video image, thereby achieving accurate determination of the expression image.
  • the determination of the superimposed position of the target matching text includes:
  • the largest inscribed graphic area is used as the superimposed position of the target matching text.
  • the embodiment of the present application detects the background area in the emoticon map; takes the largest inscribed graphic area in the background area as the superposition position of the target matching text, thereby determining the superposition position of the target matching text.
  • An embodiment of the present application also provides an emoticon package generating device, which includes:
  • the associated text determination module is used to determine the associated text of the emoticon and/or the similar emoticon package of the emoticon, the associated text of the emoticon includes at least one of subject information, scene information, emotion information, action information, and connotation information ;
  • a matching text determination module configured to determine the target matching text from the associated text of the emoticon and/or the associated text of the similar emoticon package;
  • the emoticon package generating module is used to superimpose the target matching text on the emoticon graph to generate a new emoticon package.
  • the associated text determination module includes:
  • the associated text determining unit is configured to determine the associated text of the emoticon according to the associated text of the similar emoticon package.
  • the associated text determining unit is specifically configured to:
  • determining the associated text of the emoticon from the target text includes:
  • the repetitive text is filtered out from the recognition result of the emoticon and the target text to obtain the associated text of the emoticon, and the recognition result of the emoticon includes emotion information and subject information.
  • the associated text determination module includes:
  • the emoticon matching unit is used to match the image information of the emoticon and the image information of the existing emoticon package, as well as the associated text information of the matched emoticon and the associated text information of the existing emoticon;
  • the emoticon package screening unit is configured to use the existing emoticon package as the similar emoticon package if the image matching degree or the text matching degree meets the set condition.
  • the image information includes: emoticon package category information and object category information.
  • the matching text determination module includes:
  • the matching text determining unit is configured to determine the target matching text from candidate matching texts based on at least one of the frequency of use of the words, the length of the words, and the semantics of the words, the candidate matching texts including the associated text of the emoticon and/or the The associated text of similar emoticons.
  • the matching text determining unit is specifically configured to:
  • the matching text determining unit is specifically configured to:
  • the word semantics of the candidate matching text includes word information of at least one of emotional words, hot words, and entity words;
  • the target matching text is determined from the candidate matching texts.
  • the device further includes:
  • the video image extraction module is used to extract the video image including the target part of the target object from the video before determining the associated text of the emoticon and/or the similar expression package of the emoticon;
  • the image recognition module is used to recognize the speech and/or motion execution amplitude of the extracted video image
  • the emoticon determining module is used to determine the emoticon from the video image according to the recognition result.
  • the device further includes:
  • the background detection module is used to detect the background area in the emoticon
  • the area determination module is used to determine the largest inscribed graphic area in the background area
  • the position determining module is configured to use the largest inscribed graphic area as the superimposed position of the target matching text.
  • An embodiment of the present application also provides an electronic device, which includes:
  • At least one processor At least one processor
  • a memory communicatively connected with the at least one processor; wherein,
  • the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor, so that the at least one processor can execute the instructions described in any one of the embodiments of the present application. method.
  • the embodiment of the present application also provides a non-transitory computer-readable storage medium storing computer instructions, which are used to make the computer execute the method described in any one of the embodiments of the present application.
  • FIG. 1 is a flowchart of a method for generating emoticons provided by the first embodiment of the present application
  • FIG. 2 is a flowchart of a method for generating emoticons provided by the second embodiment of the present application
  • FIG. 3 is a schematic structural diagram of an emoticon package generating apparatus provided by a fourth embodiment of the present application.
  • Fig. 4 is a block diagram of an electronic device used to implement the emoticon package generation method of an embodiment of the present application.
  • Fig. 1 is a flowchart of a method for generating emoticons provided by the first embodiment of the present application. This embodiment is applicable to the case of automatically generating emoticons.
  • the method can be executed by an emoticon package generating device.
  • the device can be implemented by software and/or hardware.
  • the emoticon package generation method provided by the embodiment of the present application includes:
  • S110 Determine the associated text of the emoticon and/or the similar emoticon package of the emoticon.
  • the emoticon is an emoticon image in an emoticon package, and the emoticon image may be a static image or a dynamic image.
  • the associated text of the emoticon may be a label of the emoticon.
  • the associated text of the specific emoticon includes at least one of the subject information, scene information, emotion information, action information, and connotation information of the emoticon.
  • the main body information includes at least one of runaway comic information, cute pet information, cute baby information, film and television variety show information, celebrity internet celebrity information, original design information, and game information.
  • the scene information includes at least one of comment information, office information, holiday events, chat information, fight picture information, love information, and education information.
  • the emotion information includes at least one of positive emotion information, neutral emotion information, and negative emotion information.
  • the action information includes at least one of entertainment information, daily behavior information, and expression information.
  • the connotation information includes various laughter information.
  • the similar emoticon pack of emoticons refers to existing emoticons that are similar to emoticons.
  • determining the associated text of the emoticon includes:
  • the recognition result of the emoticon is used as the associated text of the emoticon.
  • the recognition results of the emoticon can include emotion recognition results, subject recognition results, scene recognition results, and so on.
  • the dimensions of the associated text of the emoticon are at least two.
  • determine the similar emoticon package of the emoticon including:
  • the similar emoticon package is determined from the existing emoticon package.
  • determine the similar emoticon package of the emoticon including:
  • the similar emoticon package is determined from the existing emoticon package.
  • the image information refers to the information recognized from the image.
  • the determination of the similar emoticons of emoticons includes:
  • the existing emoticon package is used as the similar emoticon package.
  • the embodiment of the present application determines similar emoticons from the two information dimensions of image information and associated text information, thereby improving the recall rate of similar emoticons, and then improving the evaluation of similar emoticons based on the associated text of similar emoticons.
  • the recall rate of the candidate matching text of the emoticon is improved.
  • the image information includes: emoticon category information and object category information.
  • the emoticon package category information refers to information that identifies the category characteristics of the emoticon package, and may specifically be information in the label system of the emoticon package.
  • Object category information refers to the information that identifies the characteristics of the object category, which can be specifically the information in the ImageNet tag system.
  • S120 Determine the target matching text from the associated text of the emoticon and/or the associated text of the similar emoticon package.
  • the target matching text is the text that will be superimposed on the emoticon to generate an emoticon package.
  • the determining the target matching text from the associated text of the emoticon and/or the associated text of the similar emoticon package includes:
  • the target matching text is determined from candidate matching texts, the candidate matching texts including the associated text of emoticons and/or the association of the similar emoticons text.
  • the usage frequency of words refers to the frequency of occurrence of words in the associated text of the emoticon graph and/or the associated text of the similar emoticon package.
  • the determining the target matching text from the candidate matching text based on word semantics includes:
  • the word semantics of the candidate matching text includes word information of at least one of emotional words, hot words, and entity words;
  • the target matching text is determined from the candidate matching texts.
  • the determining the target matching text from the candidate matching texts based on the word length includes:
  • the target matching text is determined from candidate matching texts, including:
  • the determination of the superimposed position of the target matching text includes:
  • the largest inscribed graphic area is used as the superimposed position of the target matching text.
  • the related text of the emoticon and/or the similar emoticon package of the emoticon are determined, and the related text of the emoticon and/or the related text of the similar emoticon package are used as candidate matching texts to realize the candidate matching text Is automatically determined.
  • the target matching text By determining the target matching text from the candidate matching text, because the associated text of the emoticon includes at least one of subject information, scene information, emotion information, action information, and connotation information, and the similar emoticon package and emoticon have similar characteristics, so The determined candidate matching text describes the content information of the emoticon. Therefore, compared to determining the target matching text from the preset text, the target matching text determined in the embodiment of the present application can describe the emoticon more accurately, that is, the accuracy of the target matching text is improved. Finally, by superimposing the accurately determined target matching text on the emoticon, the automatic and accurate generation of emoticons is realized.
  • the method further includes:
  • an emoticon is determined from the video image.
  • the embodiment of the present application recognizes the speech and/or action execution amplitude of the extracted video image; according to the recognition result, the expression image is determined from the video image, thereby achieving accurate determination of the expression image.
  • the emoticon can be determined from the video image based on the recognition result of other dimensions of the video image, or on the basis of the above-mentioned recognition result, combined with the recognition result of other dimensions, so as to improve the accuracy of determining the emoticon.
  • Fig. 2 is a flowchart of a method for generating an emoticon package provided by the second embodiment of the present application. This embodiment is an optional solution proposed on the basis of the foregoing embodiment.
  • the emoticon package generation method provided by the embodiment of the present application includes:
  • S210 Determine the associated text of the emoticon according to the associated text of the similar emoticon package.
  • the determining the associated text of the emoticon according to the associated text of the similar emoticon package includes:
  • the associated text of the similar emoticon package is used as the associated text of the emoticon.
  • the determining the associated text of the emoticon according to the associated text of the similar emoticon package includes:
  • the frequency of word usage refers to the frequency of words appearing in related texts of similar emoticons.
  • the determining the associated text of the emoticon according to the associated text of the similar emoticon package includes:
  • the repetitive text is filtered out from the recognition result of the emoticon and the target text to obtain the associated text of the emoticon.
  • the determining the associated text of the emoticon according to the associated text of the similar emoticon package includes:
  • the recognition result of the emoticon includes emotion information and subject information.
  • the embodiment of the present application integrates the recognition result of the emoticon map with the target text, and uses the fused text as the associated text of the emoticon map to enrich the text content of the associated text.
  • the associated text determined based on the recognition result of the emoticon includes at least two dimensions of information, that is, the text dimension of the associated text is enriched. Therefore, based on the determination method of the embodiment of the present application, the associated text of the determined emoticon is richer and diversified.
  • S220 Determine the target matching text from the associated text of the emoticon.
  • the associated text of the emoticon is determined according to the associated text of the similar emoticon package, thereby enriching the determined dimension of the associated text of the emoticon.
  • the emoticon package generation method provided by the embodiment of the present application includes:
  • the target matching text is superimposed on the superimposed position in the target emoticon graph to generate a new emoticon package.
  • the determination of the superimposed position includes:
  • the lowermost area of the emoticon image is used as the superimposing position.
  • determining the font size of the target matching text includes:
  • determining the superimposed position of the target matching text includes:
  • the superimposed position is determined.
  • the superimposed position of the text changes with the actual situation of the picture.
  • the principle of superposition is: the superimposed text should not obscure the main body, be centered, and have a gap from the picture frame; according to the principle of not obstructing important subjects, the text can be moved to the left and right appropriately.
  • clustering algorithms are used to detect foreground and background, and morphological image processing methods such as corrosion and expansion are used to eliminate abnormal points.
  • morphological image processing methods such as corrosion and expansion are used to eliminate abnormal points.
  • the background area the largest inscribed rectangle is selected as the superimposing position. If the found position area is too small or does not exist, the lowermost area of the picture is used as the superimposing position.
  • a multi-frame foreground fusion strategy is added, which is to use the intersection of the selected fitting areas of the multi-frame images of the moving image as the final moving image sticker The bonding area, thereby improving the reliability of the selection of the bonding area.
  • determine the font color and font type including:
  • the font color and font type are determined based on at least one of edge information, color information, and angle information in the emoticon.
  • determining the font color and font type includes:
  • the emoticon is black and white
  • the font type is: random selection of Founder Ming Stone Simplified, Founder Peugeot Simplified, Founder Toot Black Simplified, Founder Fat Baby Simplified, Founder Amber Simplified, Founder Big Black Simplified, Founder Thick Round Simplified ,
  • the font color is: random black and white fonts or white and black fonts;
  • the emoticon is a color animation
  • determine the font type as follows: Randomly select Founder Ming Stone Simplified, Founder Peugeot Simplified, Founder Toot Black Simplified, Founder Fat Baby Simplified, Founder Amber Simplified, Founder Big Black Simplified, Founder Thick Round Simplified ,
  • the font color is: if the picture background is a light background, the text is colored and black, and the font color is colored (black, white, yellow, pink, blue, orange, cyan, green), and the selected color is not the same as the main picture
  • the color contrast is large, which is convenient for highlighting the font; if the background of the picture is a dark background, the text is colored with white borders, and the font color is colored (black, white, yellow, pink, blue, orange, cyan, green), and the selected color is different
  • the main color contrast with the picture is large, easy to highlight the font.
  • dynamic effects can also be randomly matched and displayed.
  • the dynamic effects include static font fitting, font scrolling carousel, and font sequential display.
  • High production efficiency The machine can instantly produce large-scale emoji resources at the first time the video source content is provided, which greatly improves the efficiency and effectiveness of emoji production;
  • High production timeliness due to machine production, it can be awakened at any time to produce at any time, processing sudden content more quickly, combined with the efficient production capacity of the machine, hot content can be quickly processed, and corresponding expression resources can be generated for users to use.
  • High-quality effect Through accurate extraction of emoticons and accurate determination of target matching text, the final emoticon package can be presented with a fake effect, making it difficult to distinguish the difference between human and machine production.
  • Fig. 3 is a schematic structural diagram of an emoticon package generating apparatus provided by a fourth embodiment of the present application.
  • the emoticon package generating apparatus 300 provided in the embodiment of the present application includes: an associated text determining module 301, a matching text determining module 302, and an emoticon package generating module 303.
  • the associated text determining module 301 is used to determine the associated text of the emoticon and/or the similar emoticon package of the emoticon.
  • the associated text of the emoticon includes subject information, scene information, emotion information, action information, and connotation information. At least one
  • the matching text determining module 302 is configured to determine the target matching text from the associated text of the emoticon and/or the associated text of the similar emoticon package;
  • the emoticon package generating module 303 is used to superimpose the target matching text on the emoticon graph to generate a new emoticon package.
  • the related text of the emoticon and/or the similar emoticon package of the emoticon are determined, and the related text of the emoticon and/or the related text of the similar emoticon package are used as candidate matching texts to realize automatic determination of candidate matching texts.
  • the target matching text By determining the target matching text from the candidate matching text, because the associated text of the emoticon includes at least one of subject information, scene information, emotion information, action information, and connotation information, and the similar emoticon package and emoticon have similar characteristics, so The determined candidate matching text describes the content information of the emoticon. Therefore, compared to determining the target matching text from the preset text, the target matching text determined in the embodiment of the present application can describe the emoticon more accurately, that is, the accuracy of the target matching text is improved. Finally, by superimposing the accurately determined target matching text on the emoticon, the automatic and accurate generation of emoticons is realized.
  • the associated text determination module includes:
  • the associated text determining unit is configured to determine the associated text of the emoticon according to the associated text of the similar emoticon package.
  • the associated text determining unit is specifically configured to:
  • the determining the associated text of the emoticon from the target text includes:
  • the repetitive text is filtered out from the recognition result of the emoticon and the target text to obtain the associated text of the emoticon, and the recognition result of the emoticon includes emotion information and subject information.
  • the associated text determination module includes:
  • the emoticon matching unit is used to match the image information of the emoticon and the image information of the existing emoticon package, as well as the associated text information of the matched emoticon and the associated text information of the existing emoticon;
  • the emoticon package screening unit is configured to use the existing emoticon package as the similar emoticon package if the image matching degree or the text matching degree meets the set condition.
  • the image information includes: emoticon package category information and object category information.
  • the matching text determination module includes:
  • the matching text determining unit is configured to determine target matching text from candidate matching texts based on at least one of the frequency of use of the words, the length of the words, and the semantics of the words, the candidate matching texts including the associated text of the emoticon and/or the The associated text of similar emoticons.
  • the matching text determining unit is specifically configured to:
  • the matching text determining unit is specifically configured to:
  • the word semantics of the candidate matching text includes word information of at least one of emotional words, hot words, and entity words;
  • the target matching text is determined from the candidate matching texts.
  • the device further includes:
  • the video image extraction module is used for extracting a video image including the target part of the target object from the video before determining the associated text of the emoticon and/or the similar emoticon package of the emoticon;
  • the image recognition module is used to recognize the speech and/or motion execution amplitude of the extracted video image
  • the emoticon determining module is used to determine the emoticon from the video image according to the recognition result.
  • the device further includes:
  • the background detection module is used to detect the background area in the emoticon
  • the area determination module is used to determine the largest inscribed graphic area in the background area
  • the position determining module is configured to use the largest inscribed graphic area as the superimposed position of the target matching text.
  • the present application also provides an electronic device and a readable storage medium.
  • FIG. 4 it is a block diagram of an electronic device according to a method for generating an emoticon package according to an embodiment of the present application.
  • Electronic devices are intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers.
  • Electronic devices can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices.
  • the components shown herein, their connections and relationships, and their functions are merely examples, and are not intended to limit the implementation of the application described and/or required herein.
  • the electronic device includes one or more processors 401, a memory 402, and interfaces for connecting various components, including a high-speed interface and a low-speed interface.
  • the various components are connected to each other by different buses, and can be installed on a common motherboard or installed in other ways as needed.
  • the processor may process instructions executed in the electronic device, including instructions stored in or on the memory to display graphical information of the GUI on an external input/output device (such as a display device coupled to an interface).
  • an external input/output device such as a display device coupled to an interface.
  • multiple processors and/or multiple buses can be used with multiple memories and multiple memories.
  • multiple electronic devices can be connected, and each device provides part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system).
  • a processor 401 is taken as an example.
  • the memory 402 is the non-transitory computer-readable storage medium provided by this application.
  • the memory stores instructions executable by at least one processor, so that the at least one processor executes the emoticon package generation method provided in this application.
  • the non-transitory computer-readable storage medium of the present application stores computer instructions, and the computer instructions are used to make a computer execute the emoticon package generation method provided in the present application.
  • the memory 402 can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as program instructions/modules corresponding to the emoticon package generation method in the embodiment of the present application (for example, The associated text determining module 301, the matching text determining module 302, and the emoticon package generating module 303 shown in FIG. 3).
  • the processor 401 executes various functional applications and data processing of the server by running non-transitory software programs, instructions, and modules stored in the memory 402, that is, realizing the emoticon package generation method in the foregoing method embodiment.
  • the memory 402 may include a storage program area and a storage data area.
  • the storage program area may store an operating system and an application program required by at least one function; the storage data area may store data created by the use of an electronic device generated by emoticons.
  • the memory 402 may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices.
  • the memory 402 may optionally include a memory remotely provided with respect to the processor 401, and these remote memories may be connected to an emoticon package generating electronic device through a network. Examples of the aforementioned networks include, but are not limited to, the Internet, corporate intranets, blockchain networks, local area networks, mobile communication networks, and combinations thereof.
  • the electronic device of the emoticon generation method may further include: an input device 403 and an output device 404.
  • the processor 401, the memory 402, the input device 403, and the output device 404 may be connected by a bus or in other ways. In FIG. 4, the connection by a bus is taken as an example.
  • the input device 403 can receive input digital or character information, and generate key signal input related to the user settings and function control of the emoticon package generating electronic device, such as touch screen, keypad, mouse, track pad, touch pad, pointing stick, a Or multiple mouse buttons, trackballs, joysticks and other input devices.
  • the output device 404 may include a display device, an auxiliary lighting device (for example, LED), a tactile feedback device (for example, a vibration motor), and the like.
  • the display device may include, but is not limited to, a liquid crystal display (LCD), a light emitting diode (LED) display, and a plasma display. In some embodiments, the display device may be a touch screen.
  • Various implementations of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, application specific ASICs (application specific integrated circuits), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: being implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, the programmable processor It can be a dedicated or general-purpose programmable processor that can receive data and instructions from the storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device. An output device.
  • the systems and techniques described here can be implemented on a computer that has: a display device for displaying information to the user (for example, a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) ); and a keyboard and a pointing device (for example, a mouse or a trackball) through which the user can provide input to the computer.
  • a display device for displaying information to the user
  • LCD liquid crystal display
  • keyboard and a pointing device for example, a mouse or a trackball
  • Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (for example, visual feedback, auditory feedback, or tactile feedback); and can be in any form (including Acoustic input, voice input, or tactile input) to receive input from the user.
  • the systems and technologies described herein can be implemented in a computing system that includes back-end components (for example, as a data server), or a computing system that includes middleware components (for example, an application server), or a computing system that includes front-end components (for example, A user computer with a graphical user interface or a web browser through which the user can interact with the implementation of the system and technology described herein), or includes such back-end components, middleware components, Or any combination of front-end components in a computing system.
  • the components of the system can be connected to each other through any form or medium of digital data communication (for example, a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), the Internet, and blockchain networks.
  • the computer system can include clients and servers.
  • the client and server are generally far away from each other and usually interact through a communication network.
  • the relationship between the client and the server is generated by computer programs that run on the corresponding computers and have a client-server relationship with each other.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Linguistics (AREA)
  • Library & Information Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Multimedia (AREA)
  • Social Psychology (AREA)
  • Psychiatry (AREA)
  • Health & Medical Sciences (AREA)
  • Human Computer Interaction (AREA)
  • Software Systems (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • User Interface Of Digital Computer (AREA)

Abstract

一种表情包生成方法、装置、设备和介质，涉及图像处理领域，尤其涉及互联网技术。该方法包括：确定表情图的关联文本和/或表情图的相似表情包，所述表情图的关联文本包括主体信息、场景信息、情绪信息、动作信息和内涵信息中的至少一种（S110）；从表情图的关联文本和/或所述相似表情包的关联文本中确定目标匹配文本（S120）；将所述目标匹配文本叠加在表情图中，以生成新的表情包（S130）。

Description

表情包生成方法、装置、设备和介质
本申请要求在2020年2月28日提交中国专利局、申请号为202010128305.6的中国专利申请的优先权，该申请的全部内容通过引用结合在本申请中。
技术领域
本申请实施例涉及图像处理领域，例如涉及互联网技术。具体地，本申请实施例提供一种表情包生成方法、装置、设备和介质。
背景技术
近年来，伴随即时通信社交的普及，互联网中用户原创内容的不断丰富，文字输入场景下的富媒体化趋势愈发明显。而表情包作为一种特殊的图像语言，被广泛使用在输入场景下。
目前为互联网用户提供表情资源的产品主要是表情搜索，即用户可手动输入文字，系统会自动为之匹配相关的表情包。
然而，作为内容型产品，表情搜索的结果强依赖于表情包资源，而目前表情资源包均靠人工制作生产，生产周期较长且成本较大。
发明内容
本申请实施例提供一种表情包生成方法、装置、设备和介质，以实现表情包的自动准确生成。
本申请实施例提供一种表情包生成方法，该方法包括：
确定表情图的关联文本和/或表情图的相似表情包，所述表情图的关联文本包括主体信息、场景信息、情绪信息、动作信息和内涵信息中的至少一种；
从表情图的关联文本和/或所述相似表情包的关联文本中确定目标匹配 文本；
将所述目标匹配文本叠加在表情图中，以生成新的表情包。
本申请实施例通过确定表情图的关联文本和/或表情图的相似表情包，将表情图的关联文本和/或相似表情包的关联文本作为候选匹配文本，以实现候选匹配文本的自动确定。
通过从候选匹配文本中确定目标匹配文本，因为表情图的关联文本包括主体信息、场景信息、情绪信息、动作信息和内涵信息中的至少一种，且相似表情包与表情图具有相似特征，所以确定的候选匹配文本描述有表情图的内容信息。因此，相比从预设文本中确定目标匹配文本，本申请实施例确定的目标匹配文本更能准确地描述表情图，也即提高了目标匹配文本的准确率。最后通过将准确确定的目标匹配文本叠加在表情图中，从而实现表情包的自动准确生成。
在阅读并理解了附图和详细描述后，可以明白其他方面。
进一步地，所述确定表情图的关联文本，包括：
根据所述相似表情包的关联文本，确定表情图的关联文本。
基于该技术特征，本申请实施例通过根据相似表情包的关联文本，确定表情图的关联文本，从而丰富表情图关联文本的确定维度。
进一步地，所述根据所述相似表情包的关联文本，确定表情图的关联文本，包括：
基于词语的使用频次，从所述相似表情包的关联文本中确定目标文本；
从所述目标文本中确定表情图的关联文本。
基于该技术特征，本申请实施例通过基于词语的使用频次，从相似表情包的关联文本中确定目标文本；从目标文本中确定表情图的关联文本，从而提高表情图关联文本的准确率。
进一步地，所述从所述目标文本中确定表情图的关联文本，包括：
匹配表情图的识别结果和所述目标文本；
根据匹配结果，确定重复文本；
从表情图的识别结果和所述目标文本中滤除所述重复文本，得到表情图的关联文本，所述表情图的识别结果包括情绪信息和主体信息。
基于该技术特征，本申请实施例通过对表情图的识别结果和目标文本进 行融合，将融合后的文本作为表情图的关联文本，以丰富关联文本的文本内容。
因为表情图的识别结果包括情绪信息和主体信息，所以基于表情图的识别结果确定的关联文本包括至少两个维度的信息，也即丰富了关联文本的文本维度。因此，基于本申请实施例的确定方式，使得确定的表情图的关联文本更加丰富多样。
进一步地，所述确定表情图的相似表情包，包括：
匹配表情图的图像信息和已有表情包的图像信息，以及匹配表情图的关联文本信息和已有表情包的关联文本信息；
若图像匹配度或文本匹配度满足设定条件，则将该已有表情包作为所述相似表情包。
基于该技术特征，本申请实施例通过从图像信息和关联文本信息两个信息维度，进行相似表情包的确定，从而提高对相似表情包的召回率，进而基于相似表情包的关联文本，提高对表情图的候选匹配文本的召回率。
进一步地，所述图像信息包括：表情包类别信息和物体类别信息。
基于该技术特征，本申请实施例通过基于表情包类别信息和物体类别信息，从已有表情包中确定相似表情包，从而提高相似表情包的准确率。
进一步地，所述从表情图的关联文本和/或所述相似表情包的关联文本中确定目标匹配文本，包括：
基于词语的使用频次、词语长度和词语语义中的至少一种，从候选匹配文本中确定目标匹配文本，所述候选匹配文本包括表情图的关联文本和/或所述相似表情包的关联文本。
基于该技术特征，本申请实施例通过基于词语的使用频次、词语长度和词语语义中的至少一种，从候选匹配文本中确定目标匹配文本，从而进一步提高目标匹配文本的确定准确率。
进一步地，所述基于词语长度，从候选匹配文本中确定目标匹配文本，包括：
若候选匹配文本的词语长度属于设定长度范围，则确定所述候选匹配文本为目标匹配文本。
基于该技术特征，本申请实施例通过判断候选匹配文本的词语长度是否属于设定长度范围，从候选匹配文本中确定目标匹配文本，从而实现目标匹 配文本的准确确定。
进一步地，所述基于词语语义，从候选匹配文本中确定目标匹配文本，包括：
确定候选匹配文本的词语语义中包括情绪词、热词和实体词中至少一种的词语信息；
根据所述词语信息，从候选匹配文本中确定目标匹配文本。
基于该技术特征，本申请实施例通过根据候选匹配文本的词语语义中包括情绪词、热词和实体词中至少一种的语义信息，从候选匹配文本中确定目标匹配文本，从而实现目标匹配文本的准确确定。
进一步地，所述确定表情图的关联文本和/或表情图的相似表情包之前，所述方法还包括：
从视频中提取包括目标对象的目标部位的视频图像；
对提取的视频图像进行说话和/或动作执行幅度的识别；
根据识别结果，从所述视频图像中确定表情图。
基于该技术特征，本申请实施例通过对提取的视频图像进行说话和/或动作执行幅度的识别；根据识别结果，从视频图像中确定表情图，从而实现表情图的准确确定。
进一步地，所述目标匹配文本的叠加位置的确定包括：
检测表情图中的背景区域；
确定背景区域中的最大内接图形区域；
将所述最大内接图形区域作为所述目标匹配文本的叠加位置。
基于该技术特征，本申请实施例通过检测表情图中的背景区域；将背景区域中的最大内接图形区域作为目标匹配文本的叠加位置，从而确定目标匹配文本的叠加位置。
本申请实施例还提供了一种表情包生成装置，该装置包括：
关联文本确定模块，用于确定表情图的关联文本和/或表情图的相似表情包，所述表情图的关联文本包括主体信息、场景信息、情绪信息、动作信息和内涵信息中的至少一种；
匹配文本确定模块，用于从表情图的关联文本和/或所述相似表情包的关联文本中确定目标匹配文本；
表情包生成模块，用于将所述目标匹配文本叠加在表情图中，以生成新 的表情包。
进一步地，所述关联文本确定模块，包括：
关联文本确定单元，用于根据所述相似表情包的关联文本，确定表情图的关联文本。
进一步地，所述关联文本确定单元，具体用于：
基于词语的使用频次，从所述相似表情包的关联文本中确定目标文本；
从所述目标文本中确定表情图的关联文本。
进一步地，所述从所述目标文本中确定表情图的关联文本，包括：
匹配表情图的识别结果和所述目标文本；
根据匹配结果，确定重复文本；
从表情图的识别结果和所述目标文本中滤除所述重复文本，得到表情图的关联文本，所述表情图的识别结果包括情绪信息和主体信息。
进一步地，所述关联文本确定模块，包括：
表情图匹配单元，用于匹配表情图的图像信息和已有表情包的图像信息，以及匹配表情图的关联文本信息和已有表情包的关联文本信息；
表情包筛选单元，用于若图像匹配度或文本匹配度满足设定条件，则将该已有表情包作为所述相似表情包。
进一步地，所述图像信息包括：表情包类别信息和物体类别信息。
进一步地，所述匹配文本确定模块，包括：
匹配文本确定单元，用于基于词语的使用频次、词语长度和词语语义中的至少一种，从候选匹配文本中确定目标匹配文本，所述候选匹配文本包括表情图的关联文本和/或所述相似表情包的关联文本。
进一步地，所述匹配文本确定单元，具体用于：
若候选匹配文本的词语长度属于设定长度范围，则确定所述候选匹配文本为目标匹配文本。
进一步地，所述匹配文本确定单元，具体用于：
确定候选匹配文本的词语语义中包括情绪词、热词和实体词中至少一种的词语信息；
根据所述词语信息，从候选匹配文本中确定目标匹配文本。
进一步地，所述装置还包括：
视频图像提取模块，用于确定表情图的关联文本和/或表情图的相似表 情包之前，从视频中提取包括目标对象的目标部位的视频图像；
图像识别模块，用于对提取的视频图像进行说话和/或动作执行幅度的识别；
表情图确定模块，用于根据识别结果，从所述视频图像中确定表情图。
进一步地，所述装置还包括：
背景检测模块，用于检测表情图中的背景区域；
区域确定模块，用于确定背景区域中的最大内接图形区域；
位置确定模块，用于将所述最大内接图形区域作为所述目标匹配文本的叠加位置。
本申请实施例还提供了一种电子设备，该设备包括：
至少一个处理器；以及
与所述至少一个处理器通信连接的存储器；其中，
所述存储器存储有可被所述至少一个处理器执行的指令，所述指令被所述至少一个处理器执行，以使所述至少一个处理器能够执行本申请实施例中任一项所述的方法。
本申请实施例还提供了一种存储有计算机指令的非瞬时计算机可读存储介质，所述计算机指令用于使所述计算机执行本申请实施例中任一项所述的方法。
附图说明
附图用于更好地理解本方案，不构成对本申请的限定。其中：
图1是本申请第一实施例提供的一种表情包生成方法的流程图；
图2是本申请第二实施例提供的一种表情包生成方法的流程图；
图3是本申请第四实施例提供的一种表情包生成装置的结构示意图；
图4是用来实现本申请实施例的表情包生成方法的电子设备的框图。
具体实施方式
以下结合附图对本申请的示范性实施例做出说明，其中包括本申请实施例的各种细节以助于理解，应当将它们认为仅仅是示范性的。因此，本领域普通技术人员应当认识到，可以对这里描述的实施例做出各种改变和修改，而不会背离本申请的范围和精神。同样，为了清楚和简明，以下的描述中省略了对公知功能和结构的描述。
第一实施例
图1是本申请第一实施例提供的一种表情包生成方法的流程图。本实施例可适用于自动生成表情包的情况。该方法可以由一种表情包生成装置来执行。该装置可以由软件和/或硬件的方式实现。参见图1，本申请实施例提供的表情包生成方法包括：
S110、确定表情图的关联文本和/或表情图的相似表情包。
其中，表情图是表情包中的表情图像，该表情图像可以是静态图像，也可以是动态图像。
具体地，表情图的关联文本可以是表情图的标签。具体表情图的关联文本包括表情图的主体信息、场景信息、情绪信息、动作信息和内涵信息中的至少一种信息。
主体信息包括暴走漫画信息、萌宠信息、萌娃信息、影视综艺信息、明星网红信息、原创设计信息和游戏信息中的至少一种。
场景信息包括评论信息、办公信息、节日赛事、聊天信息、斗图信息、恋爱信息和教育信息中的至少一种。
情绪信息包括积极情绪信息、中性情绪信息和消极情绪信息中的至少一种。
动作信息包括娱乐信息、日常行为信息和表达信息中的至少一种。
内涵信息包括各种笑点信息。
表情图的相似表情包是指与表情图相似的已有表情包。
具体地，确定表情图的关联文本，包括：
将表情图的识别结果作为表情图的关联文本。
表情图的识别结果可以包括情绪识别结果、主体识别结果和场景识别结果等。
典型地，表情图的关联文本的维度为至少两个。
具体地，确定表情图的相似表情包，包括：
匹配表情图的关联文本和已有表情包的关联文本；
根据匹配结果，从已有表情包中确定所述相似表情包。
可选地，确定表情图的相似表情包，包括：
匹配表情图的图像信息和已有表情包的图像信息；
根据匹配结果，从已有表情包中确定所述相似表情包。
其中，图像信息是指从图像中识别得到的信息。
为提高对相似表情包的召回率，所述确定表情图的相似表情包，包括：
匹配表情图的图像信息和已有表情包的图像信息，以及匹配表情图的关联文本信息和已有表情包的关联文本信息；
若图像匹配度或文本匹配度满足设定条件，则将该已有表情包作为所述相似表情包。
基于该技术特征，本申请实施例通过从图像信息和关联文本信息两个信息维度，进行相似表情包的确定，从而提高对相似表情包的召回率，进而基于相似表情包的关联文本，提高对表情图的候选匹配文本的召回率。
为提高相似表情包的准确率，图像信息包括：表情包类别信息和物体类别信息。
其中，表情包类别信息是指标识表情包类别特征的信息，具体可以是表情包的标签体系中的信息。
物体类别信息是指标识物体类别特征的信息，具体可以是ImageNet的标签体系中的信息。
S120、从表情图的关联文本和/或所述相似表情包的关联文本中确定目标匹配文本。
其中，目标匹配文本是即将叠加在表情图中以生成表情包的文本。
具体地，所述从表情图的关联文本和/或所述相似表情包的关联文本中确定目标匹配文本，包括：
基于词语的使用频次、词语长度和词语语义中的至少一种，从候选匹候选匹配文本中确定目标匹配文本，所述候选匹配文本包括表情图的关联文本和/或所述相似表情包的关联文本。
其中，词语的使用频次是指词语在表情图的关联文本和/或所述相似表情包的关联文本中的出现频率。
具体地，所述基于词语语义，从候选匹配文本中确定目标匹配文本，包括：
确定候选匹配文本的词语语义中包括情绪词、热词和实体词中至少一种的词语信息；
根据所述词语信息，从候选匹配文本中确定目标匹配文本。
具体地，所述基于词语长度，从候选匹配文本中确定目标匹配文本，包 括：
若候选匹配文本的词语长度属于设定长度范围，则确定所述候选匹配文本为目标匹配文本。
具体地，基于词语的使用频次、词语长度和词语语义，从候选匹配文本中确定目标匹配文本，包括：
基于词语的使用频次、词语长度和词语语义，分别对候选匹配文本打分；
对各打分结果进行加权求和，根据求和结果从候选匹配文本中确定目标匹配文本。
S130、将所述目标匹配文本叠加在表情图中，以生成新的表情包。
具体地，所述目标匹配文本的叠加位置的确定包括：
检测表情图中的背景区域；
确定背景区域中的最大内接图形区域；
将所述最大内接图形区域作为所述目标匹配文本的叠加位置。
本申请实施例的技术方案，通过确定表情图的关联文本和/或表情图的相似表情包，将表情图的关联文本和/或相似表情包的关联文本作为候选匹配文本，以实现候选匹配文本的自动确定。
通过从候选匹配文本中确定目标匹配文本，因为表情图的关联文本包括主体信息、场景信息、情绪信息、动作信息和内涵信息中的至少一种，且相似表情包与表情图具有相似特征，所以确定的候选匹配文本描述有表情图的内容信息。因此，相比从预设文本中确定目标匹配文本，本申请实施例确定的目标匹配文本更能准确地描述表情图，也即提高了目标匹配文本的准确率。最后通过将准确确定的目标匹配文本叠加在表情图中，从而实现表情包的自动准确生成。
进一步地，所述确定表情图的关联文本和/或表情图的相似表情包之前，所述方法还包括：
从视频中提取包括目标对象的目标部位的视频图像；
对提取的视频图像进行说话和/或动作执行幅度的识别；
根据识别结果，从所述视频图像中确定表情图。
基于该技术特征，本申请实施例通过对提取的视频图像进行说话和/或动作执行幅度的识别；根据识别结果，从视频图像中确定表情图，从而实现表情图的准确确定。
可选地，还可以根据视频图像其他维度的识别结果，或在上述识别结果的基础上，结合其他维度的识别结果，从视频图像中确定表情图，以提高表情图的确定准确率。
第二实施例
图2是本申请第二实施例提供的一种表情包生成方法的流程图。本实施例是在上述实施例的基础上提出的一种可选方案。参加图2，本申请实施例提供的表情包生成方法包括：
S210、根据所述相似表情包的关联文本，确定表情图的关联文本。
具体地，所述根据所述相似表情包的关联文本，确定表情图的关联文本，包括：
将所述相似表情包的关联文本作为表情图的关联文本。
为进一步提高表情图的关联文本的准确率，所述根据所述相似表情包的关联文本，确定表情图的关联文本，包括：
基于词语的使用频次，从所述相似表情包的关联文本中确定目标文本；
从所述目标文本中确定表情图的关联文本。
其中，词语的使用频次是指词语在相似表情包的关联文本中的出现频率。
为进一步丰富表情图的关联文本的文本内容，所述根据所述相似表情包的关联文本，确定表情图的关联文本，包括：
匹配表情图的识别结果和所述目标文本；
根据匹配结果，确定重复文本；
从表情图的识别结果和所述目标文本中滤除所述重复文本，得到表情图的关联文本。
为进一步丰富表情图的关联文本的文本维度，所述根据所述相似表情包的关联文本，确定表情图的关联文本，包括：
所述表情图的识别结果包括情绪信息和主体信息。
基于该技术特征，本申请实施例通过对表情图的识别结果和目标文本进行融合，将融合后的文本作为表情图的关联文本，以丰富关联文本的文本内容。
因为表情图的识别结果包括情绪信息和主体信息，所以基于表情图的识别结果确定的关联文本包括至少两个维度的信息，也即丰富了关联文本的文 本维度。因此，基于本申请实施例的确定方式，使得确定的表情图的关联文本更加丰富多样。
S220、从表情图的关联文本中确定目标匹配文本。
S230、将所述目标匹配文本叠加在表情图中，以生成新的表情包。
本申请实施例的技术方案，通过根据相似表情包的关联文本，确定表情图的关联文本，从而丰富表情图关联文本的确定维度。
第三实施例
本实施例是在上述实施例的基础上，以关联文本是标签为例，提出的一种可选方案。本申请实施例提供的表情包生成方法包括：
从视频资源中抽取视频图像；
对视频图像进行人脸检测，根据检测结果，从视频图像中筛选包括人脸的视频图像；
从包括人脸的视频图像中提取人脸图像；
对人脸图像进行情绪识别、是否说话识别以及是否有夸张动作识别；
根据识别结果从人脸图像中筛选符合要求的人脸图像；
对符合要求的人脸图像进行视频跟踪，确定属于同一人的人脸图像序列；
根据确定的人脸图像序列，确定静态或动态的表情图；
根据表情图的表情信息和物体信息，从已有表情包中确定第一相似表情包；
根据第一相似表情包的关联文本以及表情图中包括的明星信息、情绪信息和萌宠信息，确定表情图至少一个维度的标签；
基于表情图的标签，从已有表情包中确定第二相似表情包；
基于词语的出现频率、词语长度和词语语义，从表情图的标签、第一相似表情包的关联文本和第二相似表情包的关联文本中，确定表情图的目标匹配文本；
确定目标匹配文本中字体的大小、字体颜色、字体样式、字体动效，以及叠加位置；
根据确定的目标匹配文本中字体的大小、字体颜色、字体样式和字体动效，将目标匹配文本叠加于目标表情图中的所述叠加位置，以生成新的表情包。
具体地，叠加位置的确定包括：
检测表情图中的背景区域；
确定背景区域中最大内接图形区域；
若确定的最大内接图形区域满足设定要求，则将该区域作为叠加位置；
若确定的最大内接图形区域不满足要求，则将表情图最下方区域作为叠加位置。
具体地，确定目标匹配文本的字体大小，包括：
对目标匹配文本进行文字粒度的切分，得到文字个数；
根据文字个数、图片大小和文本的叠加位置，确定文本的字体大小。
具体地，确定目标匹配文本的叠加位置，包括：
根据表情图的内容信息，确定叠加位置。文本的叠加位置随图片真实情况改变。
叠加原则是：叠加的文本应不遮挡主体，居中，距离图片边框有空隙；可根据不遮挡重要主体的原则，将文本适当的进行向左和向右移动。
典型地，对于静图，采用聚类算法，对前景和背景进行检测，并使用腐蚀、膨胀等形态学图像处理方法，对异常点进行剔除。在背景区域，选取最大内接矩形作为叠加位置，若找到的位置区域过小或不存在，则采用图片最下方区域作为叠加位置。
对于动图，前后背景的检测及异常点处理与静图一致，在此基础上，增加多帧前景融合策略，即将动图的多帧图片所选择的贴合区域的交集作为最终的动图贴合区域，从而提高贴合区域选择可靠性。
具体地，确定字体颜色和字体类型，包括：
基于表情图中的边缘信息、颜色信息和角度信息中的至少一种，确定字体颜色和字体类型。
具体基于表情图中的边缘信息、颜色信息和角度信息中的至少一种，确定字体颜色和字体类型包括：
若表情图是为黑白图，则确定字体类型为：随机方正兰亭黑简体、方正大黑简体字体，字体颜色为：黑色或白色(图片为浅色背景选黑字，图片为深色背景选白字)；
若表情图是为彩色静图，则定字体类型为：随机选用方正铭石体简、方正标致简体、方正嘟黑简、方正胖娃简体、方正琥珀简体、方正大黑简体、 方正粗圆简体，字体颜色为：随机出黑字白边或白字黑边字体；
若表情图是为彩色动图，则确定字体类型为：随机选用方正铭石体简、方正标致简体、方正嘟黑简、方正胖娃简体、方正琥珀简体、方正大黑简体、方正粗圆简体，字体颜色为：若图片背景为浅色背景，则文字为彩色黑边，字体颜色为彩色(黑、白、黄、粉、蓝、橙、青、绿)，所选颜色别与图片的主要颜色反差较大，方便突出字体；若图片背景为深色背景，则文字为彩色白边，字体颜色为彩色(黑、白、黄、粉、蓝、橙、青、绿)，所选颜色别与图片的主要颜色反差较大，方便突出字体。
可选地，在确定字体样式后，还可以随机搭配展示动效果，动效包括字体静态贴合、字体滚动轮播和字体依次展现等。
本实施例可以产生如下技术效果：
减少人力投入：全自动化的表情资源生产，可释放生产人力，控制生产成本；
生产效率高：机器可在视频源内容提供的第一时间，瞬间生产大规模量级的表情图资源，大大提高了表情生产效率和实效性；
生产时效性高：由于机器制作，可随时唤醒随时生产，对突发内容的处理更加迅速，结合机器的高效生产能力，可对热点内容迅速处理，生成相应的表情资源供用户使用。
优质效果：通过对表情图的准确提取和目标匹配文本的准确确定，使得最终呈现的表情包可呈现以假乱真的效果，让人难以分辨人与机器生产的差别。
第四实施例
图3是本申请第四实施例提供的一种表情包生成装置的结构示意图。参见图3，本申请实施例提供的表情包生成装置300包括：关联文本确定模块301、匹配文本确定模块302和表情包生成模块303。
其中，关联文本确定模块301，用于确定表情图的关联文本和/或表情图的相似表情包，所述表情图的关联文本包括主体信息、场景信息、情绪信息、动作信息和内涵信息中的至少一种；
匹配文本确定模块302，用于从表情图的关联文本和/或所述相似表情包的关联文本中确定目标匹配文本；
表情包生成模块303，用于将所述目标匹配文本叠加在表情图中，以生 成新的表情包。
本申请实施例通过确定表情图的关联文本和/或表情图的相似表情包，将表情图的关联文本和/或相似表情包的关联文本作为候选匹配文本，以实现候选匹配文本的自动确定。
通过从候选匹配文本中确定目标匹配文本，因为表情图的关联文本包括主体信息、场景信息、情绪信息、动作信息和内涵信息中的至少一种，且相似表情包与表情图具有相似特征，所以确定的候选匹配文本描述有表情图的内容信息。因此，相比从预设文本中确定目标匹配文本，本申请实施例确定的目标匹配文本更能准确地描述表情图，也即提高了目标匹配文本的准确率。最后通过将准确确定的目标匹配文本叠加在表情图中，从而实现表情包的自动准确生成。
在一实施例中，所述关联文本确定模块，包括：
关联文本确定单元，用于根据所述相似表情包的关联文本，确定表情图的关联文本。
在一实施例中，所述关联文本确定单元，具体用于：
基于词语的使用频次，从所述相似表情包的关联文本中确定目标文本；
从所述目标文本中确定表情图的关联文本。
在一实施例中，所述从所述目标文本中确定表情图的关联文本，包括：
匹配表情图的识别结果和所述目标文本；
根据匹配结果，确定重复文本；
从表情图的识别结果和所述目标文本中滤除所述重复文本，得到表情图的关联文本，所述表情图的识别结果包括情绪信息和主体信息。
在一实施例中，所述关联文本确定模块，包括：
表情图匹配单元，用于匹配表情图的图像信息和已有表情包的图像信息，以及匹配表情图的关联文本信息和已有表情包的关联文本信息；
表情包筛选单元，用于若图像匹配度或文本匹配度满足设定条件，则将该已有表情包作为所述相似表情包。
在一实施例中，所述图像信息包括：表情包类别信息和物体类别信息。
在一实施例中，所述匹配文本确定模块，包括：
匹配文本确定单元，用于基于词语的使用频次、词语长度和词语语义中的至少一种，从候选匹配文本中确定目标匹配文本，所述候选匹配文本包括 表情图的关联文本和/或所述相似表情包的关联文本。
在一实施例中，所述匹配文本确定单元，具体用于：
若候选匹配文本的词语长度属于设定长度范围，则确定所述候选匹配文本为目标匹配文本。
在一实施例中，所述匹配文本确定单元，具体用于：
确定候选匹配文本的词语语义中包括情绪词、热词和实体词中至少一种的词语信息；
根据所述词语信息，从候选匹配文本中确定目标匹配文本。
在一实施例中，所述装置还包括：
视频图像提取模块，用于确定表情图的关联文本和/或表情图的相似表情包之前，从视频中提取包括目标对象的目标部位的视频图像；
图像识别模块，用于对提取的视频图像进行说话和/或动作执行幅度的识别；
表情图确定模块，用于根据识别结果，从所述视频图像中确定表情图。
在一实施例中，所述装置还包括：
背景检测模块，用于检测表情图中的背景区域；
区域确定模块，用于确定背景区域中的最大内接图形区域；
位置确定模块，用于将所述最大内接图形区域作为所述目标匹配文本的叠加位置。
第五实施例
根据本申请的实施例，本申请还提供了一种电子设备和一种可读存储介质。
如图4所示，是根据本申请实施例的表情包生成方法的电子设备的框图。电子设备旨在表示各种形式的数字计算机，诸如，膝上型计算机、台式计算机、工作台、个人数字助理、服务器、刀片式服务器、大型计算机、和其它适合的计算机。电子设备还可以表示各种形式的移动装置，诸如，个人数字处理、蜂窝电话、智能电话、可穿戴设备和其它类似的计算装置。本文所示的部件、它们的连接和关系、以及它们的功能仅仅作为示例，并且不意在限制本文中描述的和/或者要求的本申请的实现。
如图4所示，该电子设备包括：一个或多个处理器401、存储器402，以及用于连接各部件的接口，包括高速接口和低速接口。各个部件利用不同 的总线互相连接，并且可以被安装在公共主板上或者根据需要以其它方式安装。处理器可以对在电子设备内执行的指令进行处理，包括存储在存储器中或者存储器上以在外部输入/输出装置(诸如，耦合至接口的显示设备)上显示GUI的图形信息的指令。在其它实施方式中，若需要，可以将多个处理器和/或多条总线与多个存储器和多个存储器一起使用。同样，可以连接多个电子设备，各个设备提供部分必要的操作(例如，作为服务器阵列、一组刀片式服务器、或者多处理器系统)。图4中以一个处理器401为例。
存储器402即为本申请所提供的非瞬时计算机可读存储介质。其中，所述存储器存储有可由至少一个处理器执行的指令，以使所述至少一个处理器执行本申请所提供的表情包生成方法。本申请的非瞬时计算机可读存储介质存储计算机指令，该计算机指令用于使计算机执行本申请所提供的表情包生成方法。
存储器402作为一种非瞬时计算机可读存储介质，可用于存储非瞬时软件程序、非瞬时计算机可执行程序以及模块，如本申请实施例中的表情包生成方法对应的程序指令/模块(例如，附图3所示的关联文本确定模块301、匹配文本确定模块302和表情包生成模块303)。处理器401通过运行存储在存储器402中的非瞬时软件程序、指令以及模块，从而执行服务器的各种功能应用以及数据处理，即实现上述方法实施例中的表情包生成方法。
存储器402可以包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需要的应用程序；存储数据区可存储根据表情包生成电子设备的使用所创建的数据等。此外，存储器402可以包括高速随机存取存储器，还可以包括非瞬时存储器，例如至少一个磁盘存储器件、闪存器件、或其他非瞬时固态存储器件。在一些实施例中，存储器402可选包括相对于处理器401远程设置的存储器，这些远程存储器可以通过网络连接至表情包生成电子设备。上述网络的实例包括但不限于互联网、企业内部网、区块链网络、局域网、移动通信网及其组合。
表情包生成方法的电子设备还可以包括：输入装置403和输出装置404。处理器401、存储器402、输入装置403和输出装置404可以通过总线或者其他方式连接，图4中以通过总线连接为例。
输入装置403可接收输入的数字或字符信息，以及产生与表情包生成电子设备的用户设置以及功能控制有关的键信号输入，例如触摸屏、小键盘、 鼠标、轨迹板、触摸板、指示杆、一个或者多个鼠标按钮、轨迹球、操纵杆等输入装置。输出装置404可以包括显示设备、辅助照明装置(例如，LED)和触觉反馈装置(例如，振动电机)等。该显示设备可以包括但不限于，液晶显示器(LCD)、发光二极管(LED)显示器和等离子体显示器。在一些实施方式中，显示设备可以是触摸屏。
此处描述的系统和技术的各种实施方式可以在数字电子电路系统、集成电路系统、专用ASIC(专用集成电路)、计算机硬件、固件、软件、和/或它们的组合中实现。这些各种实施方式可以包括：实施在一个或者多个计算机程序中，该一个或者多个计算机程序可在包括至少一个可编程处理器的可编程系统上执行和/或解释，该可编程处理器可以是专用或者通用可编程处理器，可以从存储系统、至少一个输入装置、和至少一个输出装置接收数据和指令，并且将数据和指令传输至该存储系统、该至少一个输入装置、和该至少一个输出装置。
这些计算程序(也称作程序、软件、软件应用、或者代码)包括可编程处理器的机器指令，并且可以利用高级过程和/或面向对象的编程语言、和/或汇编/机器语言来实施这些计算程序。如本文使用的，术语“机器可读介质”和“计算机可读介质”指的是用于将机器指令和/或数据提供给可编程处理器的任何计算机程序产品、设备、和/或装置(例如，磁盘、光盘、存储器、可编程逻辑装置(PLD))，包括，接收作为机器可读信号的机器指令的机器可读介质。术语“机器可读信号”指的是用于将机器指令和/或数据提供给可编程处理器的任何信号。
为了提供与用户的交互，可以在计算机上实施此处描述的系统和技术，该计算机具有：用于向用户显示信息的显示装置(例如，CRT(阴极射线管)或者LCD(液晶显示器)监视器)；以及键盘和指向装置(例如，鼠标或者轨迹球)，用户可以通过该键盘和该指向装置来将输入提供给计算机。其它种类的装置还可以用于提供与用户的交互；例如，提供给用户的反馈可以是任何形式的传感反馈(例如，视觉反馈、听觉反馈、或者触觉反馈)；并且可以用任何形式(包括声输入、语音输入或者、触觉输入)来接收来自用户的输入。
可以将此处描述的系统和技术实施在包括后台部件的计算系统(例如，作为数据服务器)、或者包括中间件部件的计算系统(例如，应用服务器)、 或者包括前端部件的计算系统(例如，具有图形用户界面或者网络浏览器的用户计算机，用户可以通过该图形用户界面或者该网络浏览器来与此处描述的系统和技术的实施方式交互)、或者包括这种后台部件、中间件部件、或者前端部件的任何组合的计算系统中。可以通过任何形式或者介质的数字数据通信(例如，通信网络)来将系统的部件相互连接。通信网络的示例包括：局域网(LAN)、广域网(WAN)、互联网和区块链网络。
计算机系统可以包括客户端和服务器。客户端和服务器一般远离彼此并且通常通过通信网络进行交互。通过在相应的计算机上运行并且彼此具有客户端-服务器关系的计算机程序来产生客户端和服务器的关系。
应该理解，可以使用上面所示的各种形式的流程，重新排序、增加或删除步骤。例如，本发申请中记载的各步骤可以并行地执行也可以顺序地执行也可以不同的次序执行，只要能够实现本申请公开的技术方案所期望的结果，本文在此不进行限制。
上述具体实施方式，并不构成对本申请保护范围的限制。本领域技术人员应该明白的是，根据设计要求和其他因素，可以进行各种修改、组合、子组合和替代。任何在本申请的精神和原则之内所作的修改、等同替换和改进等，均应包含在本申请保护范围之内。

Claims (17)

  1. 一种表情包生成方法，包括：
    确定表情图的关联文本和/或表情图的相似表情包，所述表情图的关联文本包括主体信息、场景信息、情绪信息、动作信息和内涵信息中的至少一种；
    从表情图的关联文本和/或所述相似表情包的关联文本中确定目标匹配文本；
    将所述目标匹配文本叠加在表情图中，以生成新的表情包。
  2. 根据权利要求1所述的方法，其中，所述确定表情图的关联文本，包括：
    根据所述相似表情包的关联文本，确定表情图的关联文本。
  3. 根据权利要求2所述的方法，其中，所述根据所述相似表情包的关联文本，确定表情图的关联文本，包括：
    基于词语的使用频次，从所述相似表情包的关联文本中确定目标文本；
    从所述目标文本中确定表情图的关联文本。
  4. 根据权利要求3所述的方法，其中，所述从所述目标文本中确定表情图的关联文本，包括：
    匹配表情图的识别结果和所述目标文本；
    根据匹配结果，确定重复文本；
    从表情图的识别结果和所述目标文本中滤除所述重复文本，得到表情图的关联文本，所述表情图的识别结果包括情绪信息和主体信息。
  5. 根据权利要求1所述的方法，其中，所述确定表情图的相似表情包，包括：
    匹配表情图的图像信息和已有表情包的图像信息，以及匹配表情图的关联文本信息和已有表情包的关联文本信息；
    若图像匹配度或文本匹配度满足设定条件，则将该已有表情包作为所述相似表情包。
  6. 根据权利要求5所述的方法，其中，所述图像信息包括：表情包类别信息和物体类别信息。
  7. 根据权利要求1所述的方法，其中，所述从表情图的关联文本和/或所述相似表情包的关联文本中确定目标匹配文本，包括：
    基于词语的使用频次、词语长度和词语语义中的至少一种，从候选匹配文本中确定目标匹配文本，所述候选匹配文本包括表情图的关联文本和/或所述相似表情包的关联文本。
  8. 根据权利要求7所述的方法，其中，所述基于词语长度，从候选匹配文本中确定目标匹配文本，包括：
    若候选匹配文本的词语长度属于设定长度范围，则确定所述候选匹配文本为目标匹配文本。
  9. 根据权利要求7所述的方法，其中，所述基于词语语义，从候选匹配文本中确定目标匹配文本，包括：
    确定候选匹配文本的词语语义中包括情绪词、热词和实体词中至少一种的词语信息；
    根据所述词语信息，从候选匹配文本中确定目标匹配文本。
  10. 根据权利要求1所述的方法，所述确定表情图的关联文本和/或表情图的相似表情包之前，所述方法还包括：
    从视频中提取包括目标对象的目标部位的视频图像；
    对提取的视频图像进行说话和/或动作执行幅度的识别；
    根据识别结果，从所述视频图像中确定表情图。
  11. 根据权利要求1所述的方法，其中，所述目标匹配文本的叠加位置的确定，包括：
    检测表情图中的背景区域；
    确定背景区域中的最大内接图形区域；
    将所述最大内接图形区域作为所述目标匹配文本的叠加位置。
  12. 一种表情包生成装置，包括：
    关联文本确定模块，用于确定表情图的关联文本和/或表情图的相似表情包，所述表情图的关联文本包括主体信息、场景信息、情绪信息、动作信息和内涵信息中的至少一种；
    匹配文本确定模块，用于从表情图的关联文本和/或所述相似表情包的关联文本中确定目标匹配文本；
    表情包生成模块，用于将所述目标匹配文本叠加在表情图中，以生成新的表情包。
  13. 根据权利要求12所述的装置，其中，所述关联文本确定模块， 包括：
    关联文本确定单元，用于根据所述相似表情包的关联文本，确定表情图的关联文本。
  14. 根据权利要求12所述的装置，其中，所述关联文本确定模块，包括：
    表情图匹配单元，用于匹配表情图的图像信息和已有表情包的图像信息，和/或，匹配表情图的关联文本信息和已有表情包的关联文本信息；
    表情包筛选单元，用于若图像匹配度或文本匹配度满足设定条件，则将该已有表情包作为所述相似表情包。
  15. 根据权利要求12所述的装置，其中，所述匹配文本确定模块，包括：
    匹配文本确定单元，用于基于词语的使用频次、词语长度和词语语义中的至少一种，从候选匹配文本中确定目标匹配文本，所述候选匹配文本包括表情图的关联文本和/或所述相似表情包的关联文本。
  16. 一种电子设备，包括：
    至少一个处理器；以及
    与所述至少一个处理器通信连接的存储器；其中，
    所述存储器存储有可被所述至少一个处理器执行的指令，所述指令被所述至少一个处理器执行，以使所述至少一个处理器能够执行权利要求1-11中任一项所述的方法。
  17. 一种存储有计算机指令的非瞬时计算机可读存储介质，其中，所述计算机指令用于使所述计算机执行权利要求1-11中任一项所述的方法。
PCT/CN2020/100034 2020-02-28 2020-07-03 表情包生成方法、装置、设备和介质 WO2021169134A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
JP2021516910A JP7212770B2 (ja) 2020-02-28 2020-07-03 絵文字パッケージ生成方法、装置、機器および記憶媒体
KR1020217009336A KR102598496B1 (ko) 2020-02-28 2020-07-03 이모티콘 패키지 생성 방법, 장치, 설비 및 매체
US17/280,142 US11521340B2 (en) 2020-02-28 2020-07-03 Emoticon package generation method and apparatus, device and medium
EP20864289.2A EP3901786A4 (en) 2020-02-28 2020-07-03 METHOD AND DEVICE FOR MEME GENERATION AS WELL AS DEVICE AND MEDIUM

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010128305.6 2020-02-28
CN202010128305.6A CN111353064B (zh) 2020-02-28 2020-02-28 表情包生成方法、装置、设备和介质

Publications (1)

Publication Number Publication Date
WO2021169134A1 true WO2021169134A1 (zh) 2021-09-02

Family

ID=71197136

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/100034 WO2021169134A1 (zh) 2020-02-28 2020-07-03 表情包生成方法、装置、设备和介质

Country Status (3)

Country Link
EP (1) EP3901786A4 (zh)
CN (1) CN111353064B (zh)
WO (1) WO2021169134A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115223022A (zh) * 2022-09-15 2022-10-21 平安银行股份有限公司 一种图像处理方法、装置、存储介质及设备
WO2024037491A1 (zh) * 2022-08-15 2024-02-22 北京字跳网络技术有限公司 媒体内容处理方法、装置、设备及存储介质

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7212770B2 (ja) 2020-02-28 2023-01-25 ベイジン バイドゥ ネットコム サイエンス テクノロジー カンパニー リミテッド 絵文字パッケージ生成方法、装置、機器および記憶媒体
CN111353064B (zh) * 2020-02-28 2023-06-13 北京百度网讯科技有限公司 表情包生成方法、装置、设备和介质
CN112036128A (zh) * 2020-08-21 2020-12-04 百度在线网络技术（北京）有限公司 一种文本内容处理方法、装置、设备以及存储介质
CN112214632B (zh) * 2020-11-03 2023-11-17 虎博网络技术(上海)有限公司 文案检索方法、装置及电子设备
CN113239717A (zh) * 2021-02-26 2021-08-10 北京百度网讯科技有限公司 用于处理题目的方法、装置、设备、介质和程序产品
US11797153B1 (en) 2022-08-08 2023-10-24 Sony Group Corporation Text-enhanced emoji icons
CN117150063B (zh) * 2023-10-26 2024-02-06 深圳慢云智能科技有限公司 一种基于场景识别的图像生成方法及系统

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150036930A1 (en) * 2013-07-30 2015-02-05 International Business Machines Corporation Discriminating synonymous expressions using images
CN106844659A (zh) * 2017-01-23 2017-06-13 宇龙计算机通信科技(深圳)有限公司 一种多媒体数据处理方法和装置
CN107369196A (zh) * 2017-06-30 2017-11-21 广东欧珀移动通信有限公司 表情包制作方法、装置、存储介质及电子设备
CN109671137A (zh) * 2018-10-26 2019-04-23 广东智媒云图科技股份有限公司 一种图片配文字的方法、电子设备及存储介质
CN110706312A (zh) * 2019-09-20 2020-01-17 北京奇艺世纪科技有限公司 一种表情包的文案确定方法、装置及电子设备
CN110719525A (zh) * 2019-08-28 2020-01-21 咪咕文化科技有限公司 弹幕表情包的生成方法、电子设备和可读存储介质
CN111353064A (zh) * 2020-02-28 2020-06-30 北京百度网讯科技有限公司 表情包生成方法、装置、设备和介质

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
AU2011203833B2 (en) * 2010-01-11 2014-07-10 Apple Inc. Electronic text manipulation and display
JP2011192008A (ja) * 2010-03-15 2011-09-29 Zeta Bridge Corp 画像処理システムおよび画像処理方法
JP6289662B2 (ja) * 2014-07-02 2018-03-07 ホアウェイ・テクノロジーズ・カンパニー・リミテッド 情報送信方法及び送信装置
CN107239203A (zh) * 2016-03-29 2017-10-10 北京三星通信技术研究有限公司 一种图像管理方法和装置
CN108280166B (zh) * 2018-01-17 2020-01-10 Oppo广东移动通信有限公司 表情的制作方法、装置、终端及计算机可读存储介质
CN109741423A (zh) * 2018-12-28 2019-05-10 北京奇艺世纪科技有限公司 表情包生成方法及系统

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150036930A1 (en) * 2013-07-30 2015-02-05 International Business Machines Corporation Discriminating synonymous expressions using images
CN106844659A (zh) * 2017-01-23 2017-06-13 宇龙计算机通信科技(深圳)有限公司 一种多媒体数据处理方法和装置
CN107369196A (zh) * 2017-06-30 2017-11-21 广东欧珀移动通信有限公司 表情包制作方法、装置、存储介质及电子设备
CN109671137A (zh) * 2018-10-26 2019-04-23 广东智媒云图科技股份有限公司 一种图片配文字的方法、电子设备及存储介质
CN110719525A (zh) * 2019-08-28 2020-01-21 咪咕文化科技有限公司 弹幕表情包的生成方法、电子设备和可读存储介质
CN110706312A (zh) * 2019-09-20 2020-01-17 北京奇艺世纪科技有限公司 一种表情包的文案确定方法、装置及电子设备
CN111353064A (zh) * 2020-02-28 2020-06-30 北京百度网讯科技有限公司 表情包生成方法、装置、设备和介质

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3901786A4

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2024037491A1 (zh) * 2022-08-15 2024-02-22 北京字跳网络技术有限公司 媒体内容处理方法、装置、设备及存储介质
CN115223022A (zh) * 2022-09-15 2022-10-21 平安银行股份有限公司 一种图像处理方法、装置、存储介质及设备
CN115223022B (zh) * 2022-09-15 2022-12-09 平安银行股份有限公司 一种图像处理方法、装置、存储介质及设备

Also Published As

Publication number Publication date
EP3901786A1 (en) 2021-10-27
CN111353064B (zh) 2023-06-13
EP3901786A4 (en) 2021-12-01
CN111353064A (zh) 2020-06-30

Similar Documents

Publication Publication Date Title
WO2021169134A1 (zh) 表情包生成方法、装置、设备和介质
US11574470B2 (en) Suggested actions for images
US11521340B2 (en) Emoticon package generation method and apparatus, device and medium
US11481428B2 (en) Bullet screen content processing method, application server, and user terminal
US11138207B2 (en) Integrated dynamic interface for expression-based retrieval of expressive media content
US20170212892A1 (en) Predicting media content items in a dynamic interface
US11836183B2 (en) Digital image classification and annotation
JP6122499B2 (ja) 特徴に基づく候補選択
US20170083519A1 (en) Platform and dynamic interface for procuring, organizing, and retrieving expressive media content
US20170083520A1 (en) Selectively procuring and organizing expressive media content
US20210049354A1 (en) Human object recognition method, device, electronic apparatus and storage medium
US20220092071A1 (en) Integrated Dynamic Interface for Expression-Based Retrieval of Expressive Media Content
US11943181B2 (en) Personality reply for digital content
US20150067538A1 (en) Apparatus and method for creating editable visual object
CN111309200A (zh) 一种扩展阅读内容的确定方法、装置、设备及存储介质
CN111176533A (zh) 壁纸切换方法、装置、存储介质以及终端
CN111353070A (zh) 视频标题的处理方法、装置、电子设备及可读存储介质
WO2022228433A1 (zh) 信息处理方法、装置以及电子设备
CN111352685B (zh) 一种输入法键盘的展示方法、装置、设备及存储介质
US11670029B2 (en) Method and apparatus for processing character image data
CN117707370A (zh) 页面交互方法、装置、设备以及存储介质
US20210082188A1 (en) Augmented reality-based image editing
CN113485598A (zh) 聊天信息显示方法及装置
CN114205671A (zh) 基于场景对齐的视频内容剪辑方法及其装置
CN113673277A (zh) 线上绘本内容的获取方法、装置以及智能屏设备

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2021516910

Country of ref document: JP

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 20217009336

Country of ref document: KR

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 2020864289

Country of ref document: EP

Effective date: 20210324

NENP Non-entry into the national phase

Ref country code: DE