Method, electronic device, and computer program product for video processing

Patent metadata

Field	Value
Publication number	US-12185022-B2
Application number	US-202217857739-A
Country	US
Kind code	B2
Filing date	Jul 5, 2022
Priority date	Jun 10, 2022
Publication date	Dec 31, 2024
Grant date	Dec 31, 2024

Abstract

Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for video processing. The method for video processing includes receiving a first video with a first resolution from a first terminal device, the first video including an image of a first user. The method further includes converting the first video into a second video with a second resolution based on a computing model associated with the first user, the second resolution being higher than the first resolution. The method further includes sending the second video to a second terminal device. In this manner, only low-resolution videos need to be acquired and uploaded, and high-resolution videos can be obtained through reconstruction by using a computing model customized for video services, thereby ensuring that less bandwidth is occupied and improving the quality of the reconstructed videos.

First claim

What is claimed is:

1. A method for video processing, comprising:
receiving a first video with a first resolution from a first terminal device, the first video comprising an image of a first user and no other users;
converting the first video into a second video with a second resolution based on a computing model associated with the first user, the second resolution being higher than the first resolution; and
sending the second video to a second terminal device;
wherein the computing model associated with the first user comprises a first user-specific computing model customized for the first user through training that utilizes a first set of one or more additional videos each comprising one or more additional images of the first user and no other users; and
wherein the one or more additional videos collectively comprise at least a portion of a first user-specific training data set for training the first user-specific computing model, the first user-specific training data set comprising a plurality of data records each including one or more image frames at the first resolution and one or more respective corresponding image frames at the second resolution, each such image frame including an image of the same first user in a same scenario or background.

2. The method according to claim 1 , further comprising:
determining that participants of a video session on the second terminal device comprise the first user; and
obtaining the computing model associated with the first user.

3. The method according to claim 1 , wherein converting the first video into a second video with a second resolution comprises:
providing the first video to the computing model as an input to generate a video with a third resolution, the third resolution being higher than the second resolution; and
converting the video with the third resolution into the second video with the second resolution.

4. The method according to claim 1 , further comprising:
receiving a third video with a third resolution, the third video comprising the image of the first user, the third resolution being higher than the first resolution; and
generating, based on the third video, the computing model associated with the first user, the computing model being configured to convert a video comprising the image of the first user and having the first resolution into a video with the third resolution.

5. The method according to claim 4 , wherein generating the computing model associated with the first user comprises:
converting image frames in the third video into image frames with the first resolution; and
generating the computing model by using the image frames in the third video and the image frames with the first resolution after conversion as training data.

6. An electronic device, comprising:
a processor; and
a memory coupled to the processor, wherein the memory has instructions stored therein which, when executed by the processor, cause the device to execute the method of claim 1 .

7. A computer program product that is tangibly stored on a non-transitory computer-readable medium and comprises machine-executable instructions, wherein the machine-executable instructions, when executed by a machine, cause the machine to perform the method according to claim 1 .

8. The method according to claim 1 , wherein the first user is one of a plurality of participants in a video conference involving at least the first user and a second user associated with the second terminal device.

9. The method according to claim 8 , wherein the second user has an associated computing model comprising a second user-specific computing model customized for the second user through training that utilizes a second set of one or more additional videos each comprising one or more additional images of the second user.

10. A method for video processing, comprising:
receiving a first video with a first resolution from a first terminal device, the first video comprising an image of a first user and no other users;
determining a computing device near a second terminal device in video communication with the first terminal device; and
sending a computing model associated with the first user and the first video to the determined computing device, so that the computing device generates a second video with a second resolution from the first video based on the computing model, the second resolution being higher than the first resolution;
wherein the computing model associated with the first user comprises a first user-specific computing model customized for the first user through training that utilizes a first set of one or more additional videos each comprising one or more additional images of the first user and no other users; and
wherein the one or more additional videos collectively comprise at least a portion of a first user-specific training data set for training the first user-specific computing model, the first user-specific training data set comprising a plurality of data records each including one or more image frames at the first resolution and one or more respective corresponding image frames at the second resolution, each such image frame including an image of the same first user in a same scenario or background.

11. The method according to claim 10 , further comprising:
receiving a third video with a third resolution, the third video comprising the image of the first user, and the third resolution being higher than the first resolution; and
generating, based on the third video, the computing model associated with the first user, the computing model being configured to convert a video comprising the image of the first user and having the first resolution into a video with the third resolution.

12. The method according to claim 11 , wherein generating the computing model associated with the first user comprises:
converting image frames in the third video into image frames with the first resolution; and
generating the computing model by using the image frames in the third video and the image frames with the first resolution after conversion as training data.

13. An electronic device, comprising:
a processor; and
a memory coupled to the processor, wherein the memory has instructions stored therein which, when executed by the processor, cause the device to execute the method of claim 10 .

14. A computer program product that is tangibly stored on a non-transitory computer-readable medium and comprises machine-executable instructions, wherein the machine-executable instructions, when executed by a machine, cause the machine to perform the method according to claim 10 .

15. The method according to claim 10 , wherein the first user is one of a plurality of participants in a video conference involving at least the first user and a second user associated with the second terminal device.

16. The method according to claim 15 , wherein the second user has an associated computing model comprising a second user-specific computing model customized for the second user through training that utilizes a second set of one or more additional videos each comprising one or more additional images of the second user.

17. A system for video processing, comprising:
at least one computing device; and
at least one routing device;
wherein a first routing device in the at least one routing device is configured to:
receive a first video with a first resolution from a first terminal device, the first video comprising an image of a first user and no other users; and
forward the first video to a first computing device in the at least one computing device,
where

Assignees

• Dell Products Lp

Inventors

• Chen Qiang
• Fernandez Orellana Pedro

Classifications

• G06V40/10

  Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands · CPC title

• G06V10/774

  Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title

• H04N7/15

  Conference systems · CPC title

• H04N7/0117 Primary

  involving conversion of the spatial resolution of the incoming video signal (for graphics images G09G2340/0407) · CPC title

• H04N7/147 Primary

  Communication arrangements, e.g. identifying the communication as a video-communication, intermediate storage of the signals (selecting H04Q) · CPC title