WO2021107934A1 - Augmentation de la qualité d'image dans des sessions de diffusion vidéo en continu - Google Patents

Augmentation de la qualité d'image dans des sessions de diffusion vidéo en continu Download PDF

Info

Publication number
WO2021107934A1
WO2021107934A1 PCT/US2019/063344 US2019063344W WO2021107934A1 WO 2021107934 A1 WO2021107934 A1 WO 2021107934A1 US 2019063344 W US2019063344 W US 2019063344W WO 2021107934 A1 WO2021107934 A1 WO 2021107934A1
Authority
WO
WIPO (PCT)
Prior art keywords
model
face
processor
video streaming
session
Prior art date
Application number
PCT/US2019/063344
Other languages
English (en)
Inventor
Arjun Angur PATEL
Original Assignee
Hewlett-Packard Development Company, L.P.
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 Hewlett-Packard Development Company, L.P. filed Critical Hewlett-Packard Development Company, L.P.
Priority to PCT/US2019/063344 priority Critical patent/WO2021107934A1/fr
Publication of WO2021107934A1 publication Critical patent/WO2021107934A1/fr

Links

Classifications

    • 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
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/172—Classification, e.g. identification
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
    • G06F18/24133—Distances to prototypes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames

Definitions

  • Video quality during a video streaming session may be affected by a rate at which data is received for the video.
  • the rate at which data is received may be affected by the network quality of the video streaming session, which may affect the video quality of the video streaming session.
  • FIG. 1 shows a block diagram of an example apparatus that may increase an image quality of a displayed object in a video streaming session based on a determination that a network quality is below a predefined threshold;
  • FIG. 2 shows a block diagram of an example system in which the example apparatus depicted in FIG. 1 may be implemented
  • FIG. 3 shows a flow diagram of an example method for determining whether a network quality of a video streaming session is below a predefined threshold and implementing a model of an object in the video streaming session to fill-in data corresponding to missing packets to increase an image quality of the object;
  • FIG. 4 shows a block diagram of a non-transitory computer readable medium that may have stored thereon machine readable instructions for increasing an image quality of a display of a participant’s face in a telepresence session based on a model of the participant’s face.
  • the terms “a” and “an” are intended to denote at least one of a particular element.
  • the term “includes” means includes but not limited to, the term “including” means including but not limited to.
  • the term “based on” means based at least in part on.
  • Video streaming sessions to display video in real-time may include telepresence sessions, on-demand video streaming, or the like.
  • the quality of the video playback during the video streaming sessions may be dependent on the quality of the network connection.
  • a rate e.g., amount of data over time
  • the reduced rate of data transfer may enable the video streaming session to continue over degraded networks, the reduced data transfer rate may also result in reduced quality of the displayed video images, e.g., by resulting in missing pixels.
  • a processor may improve an image quality of an object being displayed in the video streaming session.
  • the video streaming session may be part of a telepresence session and the object for which the image quality is improved may be a face of a user participating in the telepresence session.
  • the processor may identify the user’s face in the telepresence session, may increase an image quality of the identified user’s face, and may display the user’s face in the telepresence session at the increased image quality.
  • the processor may increase the image quality of the user’s face in the telepresence session by regenerating missing pixels associated with the user’s face using any of various types of techniques, such as a generative adversarial network (GAN), or the like.
  • GAN generative adversarial network
  • personalized models of faces of the participants of the telepresence session may be locally stored and the processor may locally use the models (e.g., through implementation of GAN) to regenerate the missing pixels during periods of degraded network quality.
  • performance of a computing device to render video during a video streaming session may be improved.
  • the performance of the computing device to render video during the video streaming session during periods of degraded network quality may be improved.
  • improving the image quality of the video streaming session may result in reduced energy consumption, for example, by reducing a number of video streaming sessions that may be re-performed when network quality is restored.
  • FIG. 1 shows a block diagram of an example apparatus 100 that may increase an image quality of a displayed object in a video streaming session based on a determination that a network quality is below a predefined threshold.
  • FIG. 2 shows a block diagram of an example system 200 in which the example apparatus 100 depicted in FIG. 1 may be implemented. It should be understood that the example apparatus 100 depicted in FIG. 1 and the example system 200 depicted in FIG. 2 may include additional features and that some of the features described herein may be removed and/or modified without departing from the scopes of the apparatus 100 and/or the system 200.
  • the apparatus 100 may be a personal computer, a laptop computer, a tablet computer, a smartphone, a server, a node in a network (such as a data center), a network gateway, a network router, an electronic device such as Internet of Things (loT) device, a robotic device, and/or the like.
  • the apparatus 100 may include a processor 102 and a non-transitory computer readable medium, e.g., a memory 110.
  • the processor 102 may be a semiconductor-based microprocessor, a central processing unit (CPU), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), and/or other hardware device.
  • references to a single processor 102 as well as to a single memory 110 may be understood to additionally or alternatively pertain to multiple processors 102 and/or multiple memories 110.
  • the memory 110 may be an electronic, magnetic, optical, or other physical storage device that contains or stores executable instructions.
  • the memory 110 may be, for example, Read Only Memory (ROM), flash memory, solid state drive, Random Access memory (RAM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), a storage device, an optical disc, or the like.
  • the memory 110 may be a non-transitory computer readable medium. The term “non-transitory” does not encompass transitory propagating signals.
  • the processor 102 may execute instructions 112-118 to increase an image quality of a displayed object 204 in a video streaming session 202 based on an accessed model 206 of the object 204.
  • the instructions 112-118 may be machine readable instructions, e.g., non-transitory computer readable instructions.
  • the apparatus 100 may include hardware logic blocks or a combination of instructions and hardware logic blocks to implement or execute functions corresponding to the instructions 112- 118.
  • the processor 102 may fetch, decode, and execute the instructions 112 to determine whether a network quality of a video streaming session 202 is below a predefined threshold.
  • the predefined threshold may be defined as a predefined bandwidth, latency, a number of packets received over a predetermined length of time (e.g., data transmission rate), and/or the like.
  • the predefined threshold may be based on any of various manners including, for instance, a user-defined threshold value, a threshold value that is determined through testing and/or modeling, and/or the like.
  • the predefined threshold may correspond to a threshold value that may correspond to a data transmission rate that may result in a degradation of image quality of a video playback beyond a certain level, which may also be user-defined and/or determined through testing.
  • the network quality of the video streaming session 202 may be determined for a connection between devices that are participating in the video streaming session 202. That is, the processor 102 may determine the bandwidth, latency, data transmission rate, and/or the like and may compare the determined bandwidth, latency, data transmission rate, data transmission rate, and/or the like with the predefined threshold to determine whether a network quality of the video streaming session is below the predefined threshold.
  • the apparatus 100 may establish a connection with a remote apparatus 208 over the network 224 to initiate the video streaming session 202.
  • the apparatus 100 may initiate the connection using an open protocol or a proprietary protocol and the connection may be established over multiple systems, such as signaling servers, relay servers, or the like.
  • the video streaming session 202 may be a telepresence session.
  • the telepresence session may include display of the object 204 on a display of the apparatus 100, which may be an integrated display or an external display of the apparatus 100.
  • the object 204 may be an image of a participant of the telepresence session, for example, a participant co-located with the remote apparatus 208.
  • the object 204 may be a face of the participant of the telepresence session located at the remote apparatus 208 and displayed during the telepresence session at the apparatus 100. It should be understood that the object 204 may be another appropriate type of object that may be displayed during the telepresence session including, for example, an animal, an inanimate object, a background object, or the like.
  • a display quality of the object 204 may be degraded.
  • the network connection may not be able to support the video streaming session 202, and the video streaming session 202 may be terminated.
  • an amount of data transferred e.g., data transfer rate
  • the processor 102 may restore the quality of the displayed object 204 by regenerating missing data using a model 206 of the object 204.
  • each participating apparatus 100, 208 may determine whether a predefined amount of resources are available to enhance the images during the video streaming session 202.
  • the processor 102 of the apparatus 100 may determine whether the predefined amount of processing resources is available to execute instructions to regenerate missing data in the object 204 for improving a display quality of the object 204 during periods of degraded network quality.
  • the processor 102 may determine whether a remote participant’s facial model 206 is available on the apparatus 100.
  • a plurality of facial models 206 for each remote participant of the video streaming session 202 may be stored on the memory 110 or on another data store (not shown).
  • the processor 102 may request the remote participant’s facial model 206 from the remote apparatus 208. In other examples, the processor 102 may request the remote participant’s facial model 206 from the server 222.
  • each participating apparatus of the video streaming session 202 may validate a local model 210 and send the local model 210 to other participants of the video streaming session 202.
  • the apparatus 100 may first determine whether a local model 210 exists.
  • the local model 210 may be a facial model of a local participant of the video streaming session 202 and may be stored locally on the apparatus 100, remotely on a remote apparatus 208, on a remote server 222, and/or the like.
  • the apparatus 100 may determine whether the local model 210 meets a predefined criteria.
  • the processor 102 may determine whether a “loss function” associated with the local model 210 is less than a predefined value.
  • the loss function may be correlated with a similarity between the local model 210 and a current image of the participant, such that the loss function increases when the local model 210 becomes outdated.
  • the loss function may be greater than the predefined value when the participant has grown a beard since the local model 210 was originally generated. When the loss function is greater than the predefined value, a new local model 210 may be generated at the apparatus 100.
  • a new image of the participant may be captured using an interface device 212 (e.g., a camera) and a new local model 210 may be generated to overwrite the previous local model 210.
  • the processor 102 may identify the local model 210 (also referred to herein as a first model) for the face of the person participating in the telepresence session.
  • the processor 102 may determine whether an accuracy of the local model 210 to the face of the person is greater than a predetermined level. In some examples, the processor 102 may determine the accuracy of the local model 210 based on a value of the loss function.
  • the processor 102 may generate a new model 210 for the face of the person and may communicate the new model 210 to other participants of the telepresence session.
  • the processor 102 at apparatus 100 may communicate the new model 210 to the apparatus 208.
  • the remote apparatus 208 may send an updated version of its local model 210 to the apparatus 100.
  • the processor 102 may generate a new local model 210.
  • a new image of the participant may be captured using the interface device 212 to generate the new local model 210.
  • the local model 210 may be sent to the respective apparatus(es) associated with each participant of the video streaming session 202. Additionally or alternatively, the local model 210 may be stored on the server 222, and may be accessible by users from different apparatuses.
  • the apparatuses participating in the video streaming session 202 may synchronize the stored models 206 of the participants.
  • the processor 102 may send the local model 210 associated with a local participant at apparatus 100 to participating apparatuses periodically, at predetermined intervals, upon a predefined event such as generation or update of an existing model 210, or the like.
  • the apparatuses participating in the video streaming session 202 may store the models 210 on the server 222 and may synchronize the models 210 by updating the copies of models 220 at the server 222.
  • the processor 102 may determine the network quality of the video streaming session 202 between the apparatuses 100, 208 by monitoring the network 224 for various parameters including download speed, upload speed, bandwidth, latency, number of packets received, or the like.
  • the processors 102 of the apparatuses 100, 208 that are participating in the video streaming session 202 may determine the network quality of the video streaming session 202.
  • the processor 102 may share information regarding monitored network parameters with other apparatuses 208 of the video streaming session 202 to determine whether the network quality of the video streaming session is below the predefined threshold.
  • a degradation of the network quality of the video streaming session 202 may result in degraded quality of the image displayed during the video streaming session 202.
  • the degradation of the network quality may be based on a decrease in bandwidth, increased latency, reduced number of packets received, etc., between apparatuses 100, 208 participating in the video streaming session 202.
  • the network connection may not be able to support the video streaming session 202, and the video streaming session 202 may be terminated.
  • an amount of data transferred between the participating apparatuses 100 and 208 may be decreased in order to continue the video streaming session.
  • a decrease in an amount of data transferred from the remote apparatus 208 to the apparatus 100 during the video streaming session 202 may result in missing pixels in the video rendered on apparatus 100.
  • the resulting decrease in quality of the displayed object 204 may degrade the user experience, for example, if the participant’s face is not recognizable or otherwise degraded (e.g., missing pixels) during a telepresence session.
  • the processor 102 may turn on a predefined mode to initiate image compensation/enhancement during the video streaming session 202.
  • the predefined mode to initiate image compensation/enhancement may be referred to as a “super-resolution enhancement mode.”
  • a user may manually activate the predefined mode, the predefined mode may automatically be activated based on detected degradation of image quality, or the like.
  • the processor 102 may fetch, decode, and execute the instructions 114 to access the model 206 of the object 204 being displayed at apparatus 100.
  • the processor 102 may fetch, decode, and execute the instructions 116 to increase the image quality of the displayed object 204 based on the accessed model 206 of the object 204. For instance, the processor 102 may identify a group of pixels 214 corresponding to the object 204 in the displayed image. The group of pixels 214 that form the object 204 may be missing pixels.
  • the processor 102 may generate new pixels 216 for the identified group of pixels 214 based on the model 206 of the object 204 to increase the image quality of the displayed object 204 and display the object 204 with the generated new pixels 216.
  • the processor 102 may generate the new pixels 216 to fill-in the missing pixels in the identified group of pixels 214.
  • the new pixels 216 may be generated using machine learning techniques such as GAN.
  • the processor 102 may implement a neural network that is trained on the model 206 to regenerate missing data in the object 204 corresponding to the model 206.
  • the processor 102 may perform a process for primary object outlining.
  • the object 204 may be an image of a face of a participant of a telepresence session and the processor 102 may identify the object 204 based on depth sensing techniques to select a group of pixels at a particular depth as being part of the object 204. Depth sensing cameras may capture depth information for pixels in the image.
  • the processor 102 may identify the object 204 through implementation of object segmentation techniques that use machine learning to differentiate objects in an image.
  • the processor 102 may process the group of pixels 214 identified to correspond to the object 204 in order to generate the new pixels 216.
  • the processor 102 may generate the new pixels 216 based on the model 206 that is specific to the object 204.
  • the model 206 may be a facial model of a particular participant of a telepresence session. Since the model 206 is unique to the face of the particular participant, the processor 102 may regenerate the image of the particular participant’s face accurately.
  • the processor 102 may fetch, decode, and execute the instructions 118 to display the object 204 at the increased image quality.
  • the apparatus 100 may include a display to display the video streaming session 202 including the object 204.
  • the processor 102 may enhance a plurality of objects 204 during the video streaming session 202.
  • each of the plurality of objects 204 may be associated with a corresponding model 206.
  • Each of the plurality of objects 204 may be one of a participant’s face, an animal, an inanimate object, a background object, or another object that is displayed in the video streaming session 202.
  • the video streaming session 202 may be for on- demand video streaming in which a video 218 may be streamed from the server 222 on the apparatus 100.
  • the object 204 may correspond to an object that appears in the video 218 and a model 220 may be associated with the object.
  • the object 204 may be an actor’s face that appears in the video 218, and the model 220 may be a facial model of the actor.
  • the model 220 may be generated in advance and stored on the server 222 for access together with the video 218.
  • the apparatus 100 may retrieve the model 220 for local storage in the memory 110, to be used at a later time when streaming the video 218 during periods of degraded network quality.
  • the video streaming session 202 may include a plurality of objects 204 and a plurality of models 220 corresponding to the plurality of objects 204 may be pre generated.
  • the process for enhancing an object 204 in a video streaming session 202 as described herein may be implemented to reduce a size of content stored on a server 222.
  • a size of the video 218 that is stored on the server 222 may be reduced, which may result in lost pixels in the video 218.
  • the processor 102 may use the model 220 to enhance a quality of object 204 during reproduction on the apparatus 100, to fill-in the missing pixels with new pixels 216, as previously described.
  • FIG. 3 there is shown a flow diagram of an example method 300 for determining whether a network quality of a video streaming session 202 is below a predefined threshold and implementing a model 206 of an object 204 in the video streaming session 202 to fill-in data corresponding to missing packets to increase an image quality of the object 204.
  • the method 300 depicted in FIG. 3 may include additional operations and that some of the operations described therein may be removed and/or modified without departing from the scope of the method 300.
  • the description of the method 300 is also made with reference to the features depicted in FIGS. 1-2 for purposes of illustration. Particularly, the processor 102 of the apparatus 100 may execute some or all of the operations included in the method 300.
  • the processor 102 may determine whether a network quality of the video streaming session 202 is below a predefined threshold.
  • the network quality of the video streaming session 202 may be degraded due to reduced bandwidth, increased latency, or the like.
  • the network quality may be monitored by one or more of the participants of the video streaming session 202, such as the apparatus 100, a remote apparatus 208, and/or a server 222.
  • the processor 102 may identify the object 204 of the video streaming session 202 being displayed. In some examples, the video streaming session 202 may be missing packets due to the network quality being below the predefined threshold. [0043] At block 306, the processor 102 may implement a model 206 of the object 204 to fill-in data corresponding to the missing packets to increase an image quality of the object 204. At block 308, the processor 102 may display the object 204 in the video streaming session 202 at the increased image quality.
  • the processor 102 may determine whether the model 206 of the object 204 is available.
  • the model 206 of the object 204 may be stored on the memory 110 of the apparatus 100.
  • the model 206 may also be stored on an apparatus at which a participant of the video streaming session 202 associated with the object 204 is located.
  • the processor 102 may output a request for the model 206 of the object 204.
  • the processor 102 may receive the model 206 of the object 204 from the remote apparatus 208 associated with the participant corresponding to the model 206.
  • the processor 102 may also implement the received model 206 to improve a quality of the object 204 at the apparatus 100.
  • the processor 102 may identify a group of pixels 214 associated with the identified object 204 being displayed.
  • the processor 102 may generate new pixels 216 for the identified group of pixels 214 based on the model 206 of the object 204.
  • the new pixels 216 may incorporate the data corresponding to the missing packets.
  • the processor 102 may display the object 204 with the generated new pixels 216 to enhance a quality of the displayed object 204.
  • the identified object 204 may be a person’s face in the video streaming session 202 and the model 206 of the object 204 may be a model of the person’s face.
  • the video streaming session 202 may be part of a telepresence session and the identified object 204 may be an image of a face of a participant of the telepresence session.
  • the identified object 204 may be an object in the video streaming session 202 as designated by the participants of the telepresence session.
  • the video streaming session 202 may be on- demand data streaming of a video 218 from a server 222.
  • a plurality of models 220 associated with objects 204 appearing in the video 218 may be stored on the server 222.
  • the apparatus 100 may retrieve the models 220 associated with the video 218 to enhance a plurality of objects 204 in the video 218 using the retrieved models 220.
  • the models 220 may be facial models of people appearing in the video 218, and the models 220 may be pre-generated and stored on the server 222.
  • the processor 102 may identify a local model 210 (e.g., a first model) of a user’s face, the user being a local participant of the telepresence session at the apparatus 100.
  • the processor 102 may determine whether an accuracy of the local model to the user is less than a predetermined level. Based on a determination that the accuracy of the local model 210 is less than the predetermined level, the processor 102 may generate a new local model 210 of the user’s face.
  • the local model 210 may be compared with an image of a local user captured using an interface device 212.
  • the accuracy of the local model 210 may be less than the predetermined level when an appearance of the local user has changed (e.g., the user has grown a beard since the local model 210 was previously generated).
  • the new model 210 may be generated based on the image of the local user captured using the interface device 212.
  • the processor 102 may communicate the new local model 210 of the user’s face to other participants of the telepresence session.
  • Some or all of the operations set forth in the method 300 may be included as utilities, programs, or subprograms, in any desired computer accessible medium.
  • the method 300 may be embodied by computer programs, which may exist in a variety of forms both active and inactive. For example, they may exist as machine readable instructions, including source code, object code, executable code or other formats. Any of the above may be embodied on a non-transitory computer readable storage medium.
  • non-transitory computer readable storage media include computer system RAM, ROM, EPROM, EEPROM, and magnetic or optical disks or tapes. It is therefore to be understood that any electronic device capable of executing the above-described functions may perform those functions enumerated above.
  • FIG. 4 there is shown a block diagram of a non- transitory computer readable medium 400 that may have stored thereon machine readable instructions for increasing an image quality of a display of a participant’s face in a video streaming session 202 based on a model 206 of the participant’s face.
  • the video streaming session 202 may be a telepresence session.
  • the computer readable medium 400 depicted in FIG. 4 may include additional instructions and that some of the instructions described herein may be removed and/or modified without departing from the scope of the computer readable medium 400 disclosed herein.
  • the computer readable medium 400 may be a non-transitory computer readable medium.
  • non-transitory does not encompass transitory propagating signals.
  • the description of the non-transitory computer readable medium 400 is also made with reference to the features depicted in FIGS. 1-2 for purposes of illustration. Particularly, the processor 102 of the apparatus 100 may execute some or all of the instructions 402-408 included in the non-transitory computer readable medium 400.
  • the computer readable medium 400 may have stored thereon machine readable instructions 402-408 that a processor, such as the processor 102 depicted in FIGS. 1 and 2, may execute. Particularly, the processor 102 may execute instructions 402 to determine whether a network quality of a telepresence session is below a predefined threshold.
  • the processor 102 may execute the instructions 404 to identify a face of a participant of the telepresence session 202.
  • the processor 102 may execute the instructions 406 to determine whether a model 206 of the participant’s face is available on the apparatus 100. Based on a determination that the model 206 of the participant’s face is available, the processor 102 may execute the instructions 406 to increase an image quality of a display of the participant’s face based on the model 206 of the participant’s face.
  • the processor 102 may execute the instructions 408 to display the participant’s face in the telepresence session at the increased image quality.
  • the processor 102 may identify a group of pixels 214 associated with the identified face of the participant of the telepresence session and may generate new pixels 216 for the identified group of pixels 214 based on the model 206 of the participant’s face.
  • the processor 102 may also display the participant’s face in the telepresence session with the generated new pixels 216 such that the displayed face of the participant with the generated new pixels 216 may have a higher image quality than with the identified group of pixels 214.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Software Systems (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Computing Systems (AREA)
  • Databases & Information Systems (AREA)
  • Medical Informatics (AREA)
  • Human Computer Interaction (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)

Abstract

Selon des exemples, l'invention concerne un appareil qui peut comprendre un processeur et une mémoire sur laquelle peuvent être stockées des instructions qui, lorsqu'elles sont exécutées par le processeur, peuvent amener le processeur à déterminer si une qualité de réseau d'une session de diffusion vidéo en continu est inférieure à un seuil prédéfini, la session de diffusion vidéo en continu pouvant comprendre un affichage d'un objet. Sur la base d'une détermination selon laquelle la qualité de réseau est inférieure au seuil prédéfini, le processeur peut accéder à un modèle de l'objet affiché et peut augmenter une qualité d'image de l'objet affiché sur la base du modèle accédé de l'objet. Le processeur peut afficher l'objet à la qualité d'image augmentée.
PCT/US2019/063344 2019-11-26 2019-11-26 Augmentation de la qualité d'image dans des sessions de diffusion vidéo en continu WO2021107934A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/US2019/063344 WO2021107934A1 (fr) 2019-11-26 2019-11-26 Augmentation de la qualité d'image dans des sessions de diffusion vidéo en continu

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/US2019/063344 WO2021107934A1 (fr) 2019-11-26 2019-11-26 Augmentation de la qualité d'image dans des sessions de diffusion vidéo en continu

Publications (1)

Publication Number Publication Date
WO2021107934A1 true WO2021107934A1 (fr) 2021-06-03

Family

ID=76129852

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2019/063344 WO2021107934A1 (fr) 2019-11-26 2019-11-26 Augmentation de la qualité d'image dans des sessions de diffusion vidéo en continu

Country Status (1)

Country Link
WO (1) WO2021107934A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114827664A (zh) * 2022-04-27 2022-07-29 咪咕文化科技有限公司 多路直播混流方法、服务器、终端设备、系统及存储介质

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140225978A1 (en) * 2005-03-01 2014-08-14 EyesMatch Ltd. Method for image transformation, augmented reality, and teleperence
US20160352791A1 (en) * 2015-05-27 2016-12-01 Google Inc. Streaming spherical video
US20180248930A1 (en) * 2015-04-27 2018-08-30 International Business Machines Corporation Dynamically adjusting quality of service using cognitive focus of attention detection

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140225978A1 (en) * 2005-03-01 2014-08-14 EyesMatch Ltd. Method for image transformation, augmented reality, and teleperence
US20180248930A1 (en) * 2015-04-27 2018-08-30 International Business Machines Corporation Dynamically adjusting quality of service using cognitive focus of attention detection
US20160352791A1 (en) * 2015-05-27 2016-12-01 Google Inc. Streaming spherical video

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114827664A (zh) * 2022-04-27 2022-07-29 咪咕文化科技有限公司 多路直播混流方法、服务器、终端设备、系统及存储介质
CN114827664B (zh) * 2022-04-27 2023-10-20 咪咕文化科技有限公司 多路直播混流方法、服务器、终端设备、系统及存储介质

Similar Documents

Publication Publication Date Title
US10609104B2 (en) Live stream manifests for on demand content
JP6561177B2 (ja) メディア・ファイルのアップロードおよびトランスコーディング
WO2020093781A1 (fr) Procédé et appareil de détermination de taux de clics estimés de ressources multimédias et serveur
US11190566B1 (en) Generating requests for streaming media
US10938725B2 (en) Load balancing multimedia conferencing system, device, and methods
US8385425B2 (en) Temporal video filtering for real time communication systems
US11196962B2 (en) Method and a device for a video call based on a virtual image
JP2024073439A (ja) 選択的な動きの描画を伴う画像表示
US10574713B2 (en) Self-adaptive sample period for content sharing in communication sessions
US10623450B2 (en) Access to data on a remote device
JP2017519406A (ja) ネットワークビデオ再生方法及び装置
US20230379763A1 (en) Dynamic continuous quality of service adjustment system
AU2018275194A1 (en) Temporal placement of a rebuffering event
KR20230028250A (ko) 강화 학습 기반 속도 제어
CN111343503A (zh) 视频的转码方法、装置、电子设备及存储介质
WO2021107934A1 (fr) Augmentation de la qualité d'image dans des sessions de diffusion vidéo en continu
CN107277624B (zh) 时长计算方法及其装置
US8953843B1 (en) Selecting objects in a sequence of images
US8977003B1 (en) Detecting objects in a sequence of images
US10666698B1 (en) Bit rate selection for streaming media
CN108989905B (zh) 媒体流控制方法、装置、计算设备及存储介质
CN105872634A (zh) 视频数据共享的方法及装置
US20150215671A1 (en) Video sharing mechanism where in the filters can be changed after the video is shared with a filter
US11399208B2 (en) Packet priority for visual content
US10412181B2 (en) Methods and systems for adjusting user experience on a social networking service using quotas

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: 19954150

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19954150

Country of ref document: EP

Kind code of ref document: A1