Video coding method, system and server

Patent metadata

Field	Value
Publication number	US-2020107057-A1
Application number	US-201916703701-A
Country	US
Kind code	A1
Filing date	Dec 4, 2019
Priority date	Jan 14, 2015
Publication date	Apr 2, 2020
Grant date	—

Abstract

Disclosed is a video coding method performed at a video coding system, in which original collected video data is received from a coding requesting terminal; each video frame image of the original collected video data is divided into multiple subarea images; the subarea images divided from each video frame image are transmitted to coding servers of which a coder is performed from a multithreaded framework into a multi-process framework, to code a respective subset of the subarea images in a second coding scheme and output a coded video stream corresponding to the respective subset of the subarea images; and the coded video streams are then merged into a target video stream in the second coding scheme and sent to a user terminal. The user terminal decodes the target video stream from the second coding scheme into a second sequence of video frame images to be played at the user terminal.

First claim

1 . A video coding method performed at a video coding system including a distributing server, a managing server, multiple coding servers, and a stream merging server, wherein the distributing server is connected to the coding servers, the coding servers are connected to the stream merging server, and the managing server is connected to the distributing server, the coding servers and the stream merging server, respectively, each server configured to perform a respective task of the video coding method, the method comprising:
receiving, by the distributing server, original collected video data from a coding requesting terminal; dividing, by the managing server, each video frame image of the original collected video data into multiple subarea images; transmitting, by the distributing server, the subarea images divided from each video frame image of the original collected video data to the multiple coding servers of which a coder is performed from a multithreaded framework into a multi-process framework, to code the subarea images in a second coding scheme and output a coded video stream; coding, by each of the multiple coding servers, a respective subset of the subarea images in the second coding scheme and outputting, to the stream merging server, a coded video stream corresponding to the respective subset of the subarea images; merging, by the stream merging server, the coded video streams output by the multiple coding servers into a target video stream in the second coding scheme; and sending, by the stream merging server, the target video stream to a user terminal specified by the coding requesting terminal, wherein the user terminal decodes the target video stream from the second coding scheme into a second sequence of video frame images to be played at the user terminal.

2 . The method of claim 1 , further comprising:
uniformly dividing, by the managing server, each video frame image of the original collected video data into multiple subarea images, so that the divided subarea images have a same size.

3 . The method of claim 1 , wherein transmitting, by the distributing server, the subarea images divided from each video frame image of the original collected video data to the multiple coding servers of which a coder is performed from a multithreaded framework into a multi- process framework, to code the subarea images in a second coding scheme and output a coded video stream comprises:
according to a location of each of the subarea images in the video frame image, from which the subarea image is divided, assigning, by the managing server, a respective coding server address in a unit of subarea image stream data formed by the divided subarea images at a same location; and transmitting, by the distributing server, each piece of subarea image stream data to a respective one of the coding servers, corresponding to the assigned coding server address, to code the piece of subarea image stream data in the second coding scheme and output the coded video stream.

4 . The method of claim 1 , wherein transmitting, by the distributing server, the subarea images divided from each video frame image to multiple coding servers of which a coder is performed from a multithreaded framework into a multi-process framework, to code the subarea images in a second coding scheme and output a coded video stream comprises:
when a number of the subarea images divided from each video frame image of the original collected video data is equal to a number of the coding servers, distributing, by the distributing server, the subarea images divided from each video frame image to a corresponding one of the multiple coding servers, coding, by the corresponding coding server, the subarea images in the second coding scheme, and outputting the coded subarea images in sequence to form the coded video stream; and when the number of the subarea images divided from each video frame image of the original collected video data is not equal to the number of the coding servers, scheduling, by the managing server, a current least loaded coding server for the subarea images divided from each video frame image, to code the subarea images in the second coding scheme and output the coded video stream.

5 . The method of claim 1 , wherein merging, by the stream merging server, the coded video streams output by the coding servers into a target video stream in the second coding scheme comprises:
for each of the coded images in the coded video stream output by each of the multiple coding servers,
obtaining, by the stream merging server, location information of a corresponding subarea image relative to the video frame image from which the subarea image is divided; and
merging, by the stream merging server, the coded video streams output by the coding servers into the target video stream according to the location information.

6 . The method of claim 1 , further comprising:
returning, by the stream merging server, the target video stream back to the coding requesting terminal.

7 . The method of claim 1 , wherein the second coding scheme is an H.265-format coding scheme.

8 . A video coding system including a distributing server, a managing server, multiple coding servers, and a stream merging server, wherein the distributing server is connected to the coding servers, the coding servers are connected to the stream merging server, and the managing server is connected to the distributing server, the coding servers and the stream merging server, respectively, each server comprising a nonvolatile storage medium and a processor; the nonvolatile storage medium having instructions stored therein, the instructions, when executed by the processor, causing the processor to perform a respective task of a video coding method including the following steps:
receiving, by the distributing server, original collected video data from a coding requesting terminal; dividing, by the managing server, each video frame image of the original collected video data into multiple subarea images; transmitting, by the distributing server, the subarea images divided from each video frame image of the original collected video data to the multiple coding servers of which a coder is performed from a multithreaded framework into a multi-process framework, to code the subarea images in a second coding scheme and output a coded video stream; coding, by each of the multiple coding servers, a respective subset of the subarea images in the second coding scheme and outputting, to the stream merging server, a coded video stream corresponding to the respective subset of the subarea images; merging, by the stream merging server, the coded video streams output by the multiple coding servers into a target video stream in the second coding scheme; and sending, by the stream merging server, the target video stream to a user terminal specified by the coding requesting terminal, wherein the user terminal decodes the target video stream from the second coding scheme into a second sequence of video frame images to be played at the user terminal.

9 . The video coding system of claim 8 , wherein the instructions, when executed by the processor, further causes the processor to perform the following steps:
uniformly dividing, by the managing server, each video frame image of the original collected video data into multiple subarea images, so that the divided subarea images have a same size.

10 . The video coding system of claim 8 , wherein the step of transmitting, by the distributing server, the subarea images divided from each video frame image of the original collected video data to the multiple coding servers of which a coder is performed from a multithreaded framework into a multi-process framework, to code

Assignees

• Tencent Tech Shenzhen Co Ltd

Inventors

• Hu Xinghua
• Zeng Xinhai

Classifications

• H04N21/234345 Primary

  the reformatting operation being performed only on part of the stream, e.g. a region of the image or a time segment · CPC title

• H04N19/436

  using parallelised computational arrangements · CPC title

• H04L65/60

  Network streaming of media packets · CPC title

• H04N21/2343

  involving reformatting operations of video signals for distribution or compliance with end-user requests or end-user device requirements {(video transcoding H04N19/40; media packet handling at the source H04L65/762)} · CPC title