Video coding and decoding methods and apparatus

Patent metadata

Field	Value
Publication number	US-10638142-B2
Application number	US-201515569840-A
Country	US
Kind code	B2
Filing date	Dec 21, 2015
Priority date	Apr 30, 2015
Publication date	Apr 28, 2020
Grant date	Apr 28, 2020

Abstract

The present application provides a video encoding method, decoding method and devices thereof. The video encoding device includes a video capturing unit for capturing a video image; a processing unit for performing compression encoding on a background image in the video image to obtain video compression data and for structuralizing a foreground moving object in the video image to obtain foreground object metadata; a data transmission unit for transmitting the video compression data and the foreground object metadata, wherein the foreground object metadata is data that stores video structuralized semantic information. In the case that the moving objects are large in quantity or in size, the amount of video data can be effectively reduced and the limitation of network bandwidth during transmission can be mitigated.

First claim

The invention claimed is:

1. A video encoding method, comprising:
capturing a video image to be transmitted;
performing compression encoding on a background image in the video image to obtain video compression data, and structuralizing a foreground image comprising a foreground moving object in the video image to obtain foreground object metadata comprising position information and time information;
transmitting the video compression data and the foreground object metadata to a decoding device without transmitting the foreground image, so that the decoding device decodes the video compression data to obtain a background image, and interprets the foreground object metadata to obtain a foreground image, and superimposes the foreground image on a corresponding position of the background image based on the position information and time information in the metadata for synthetic display to reproduce the captured video image; wherein the foreground object metadata describes semantically the foreground object.

2. The method according to claim 1 , wherein, it further comprises:
establishing a background model of the video image and detecting the foreground moving object based on the established background model to separate the background image and the foreground moving object in a current frame of the video image.

3. The method according to claim 1 , wherein,
the video compression data corresponding to the background image is transmitted at a set time interval, and the foreground object metadata corresponding to the foreground moving object is transmitted in real time.

4. The method according to claim 1 , wherein,
the structuralization algorithm used when structuralizing the foreground moving object in the video image includes structuralization algorithms that do not set object type and structuralization algorithms that set object type.

5. A video decoding method, comprising:
receiving, by a decoding device, video compression data and foreground object metadata that describes semantically a foreground image comprising a foreground object;
decoding the video compression data to obtain a background image and semantically interpreting the foreground object metadata to reproduce the foreground image;
synthetically displaying a decoded background image and the reproduced foreground image, comprising:
superimposing the foreground image on a corresponding position of the background image based on position information and time information in the metadata for synthetic display, reproducing the captured video image.

6. The method according to claim 5 , wherein, synthetically displaying the decoded background image and the reproduced foreground image further comprises:
selecting a corresponding foreground object image from pre-stored images as the foreground moving object based on information of the foreground object metadata, and synthetically displaying the foreground object image and the decoded background image.

7. The method according to claim 5 , wherein, synthetically displaying the decoded background image and the reproduced foreground image further comprises:
superimposedly plotting the foreground moving object described by the foreground object metadata on the decoded background image by means of display plotting technique based on the information of the foreground object metadata.

8. The method according to claim 1 , wherein
capturing a video image to be transmitted comprises:
capturing a video image on a highway;
performing compression encoding on a background image in the video image to obtain video compression data and structuralizing a foreground image comprising a foreground moving object in the video image to obtain foreground object metadata comprises:
separating a frame of the video image into a background image containing a still scene and a foreground image containing a moving object vehicle according to a background model;
encoding the background image into video compression data in form of digital array through compression encoding, and structuralizing the foreground image of the moving object vehicle to obtain foreground object metadata, wherein the foreground object metadata is data that stores video structuralized semantic information;
transmitting the video compression data and the foreground object metadata without transmitting the foreground image, wherein the foreground object metadata describes semantically the foreground object and is semantically interpreted by the decoding device to reproduce the foreground image comprises:
mixing the video compression data and the foreground object metadata to obtain a mixed stream of video data with metadata and transmitting the mixed stream.

9. The method according to claim 8 , wherein,
the foreground object metadata contains at least: vehicle type, vehicle color, vehicle brand, vehicle model, license plate number, position of the foreground object in the frame of the video image, and time of the frame of the video image.

10. The method according to claim 5 , wherein, interpreting the foreground object metadata to reproduce the foreground image comprises:
selecting a corresponding foreground object image as a foreground moving object based on information of the foreground object metadata, or
superimposedly plotting a foreground moving object described by the foreground object metadata on the decoded background image by means of display plotting technique based on information of the foreground object metadata.

11. An encoding device, wherein, the encoding device comprises a processor, a memory, a communication interface and a bus;
the processor, the memory and the communication interface are connected via the bus and communicate with each other;
the memory stores executable program codes;
the processor runs a program corresponding to the executable program codes by reading the executable program codes stored in the memory for:
capturing a video image to be transmitted;
performing compression encoding on a background image in the video image to obtain video compression data, and structuralizing a foreground image comprising a foreground moving object in the video image to obtain foreground object metadata comprising position information and time information;
transmitting the video compression data and the foreground object metadata to a decoding device without transmitting the foreground image, so that the decoding device decodes the video compression data to obtain a background image, and interprets the foreground object metadata to obtain a foreground image, and superimposes the foreground image on a corresponding position of the background image based on the position information and time information in the metadata for synthetic display to reproduce the captured video image; wherein the foreground object metadata describes semantically the foreground object.

12. A decoding device, wherein, the decoding device comprises a processor, a memory, a communication interface and a bus;
the processor, the memory and the communication interface re connected via the bus and communicate with each other;
the memory stores executable program codes;
the processor runs a program corresponding to the executable program codes by reading the executable program codes stored in the memory for:
receiving, by a decoding device, video compression data and foreground object metadata that describes semantically a foreground image comprising a foreground object;
decoding the video compression data to obtain a background image and semantically interpreting the foreground object metadata to reproduce the foreground image;
synthetically displaying a decoded background image and the reproduced foreground image, co

Assignees

• Hangzhou Hikvision Digital Tec

Inventors

• Guo Bin
• Cai Weiwei

Classifications

• H04N19/463

  by compressing encoding parameters before transmission · CPC title

• H04N19/54

  using feature points or meshes · CPC title

• H04N19/139

  Analysis of motion vectors, e.g. their magnitude, direction, variance or reliability · CPC title

• H04N19/167

  Position within a video image, e.g. region of interest [ROI] · CPC title

• H04N19/27

  involving both synthetic and natural picture components, e.g. synthetic natural hybrid coding [SNHC] · CPC title