Compression of a video stream having frames with relatively heightened quality parameters on blocks on an identified point of interest (PoI)

Patent metadata

Field	Value
Publication number	US-10893281-B2
Application number	US-201816159467-A
Country	US
Kind code	B2
Filing date	Oct 12, 2018
Priority date	Oct 12, 2018
Publication date	Jan 12, 2021
Grant date	Jan 12, 2021

Abstract

A computer-implemented method according to one embodiment includes receiving specification of a type of point of interest in video data, analyzing frames in the video data for identifying a point of interest in the frames of the specified type, adjusting quality parameters on blocks on the identified point of interest for improving a quality thereof, adjusting quality parameters on blocks not on the identified point of interest for reducing a quality thereof, and outputting a compressed video stream having the adjusted quality parameters. A computer program product for compressing video data according to another embodiment includes a computer readable storage medium having program instructions embodied therewith. The computer readable storage medium is not a transitory signal per se. The program instructions are readable and/or executable by a computer to cause the computer to perform the foregoing method.

First claim

What is claimed is:

1. A computer-implemented method, comprising:
receiving specification of a type of point of interest in video data;
analyzing frames in the video data to identify a point of interest in the frames of the specified type,
wherein the frames in the video data are analyzed by an object detector;
adjusting quality parameters on blocks on the identified point of interest to improve a quality thereof,
wherein a first of the quality parameters adjusted on blocks on the identified point of interest includes motion estimation,
wherein adjusting the motion estimation includes increasing an amount of motion estimation bits on blocks on the identified point of interest;
adjusting quality parameters on blocks not on the identified point of interest to reduce a quality thereof; and
outputting a compressed video stream having the adjusted quality parameters.

2. The computer-implemented method of claim 1 , wherein the quality parameters adjusted on blocks on the identified point of interest are additionally selected from the group consisting of: quantization values of the blocks, number of P-frames, and group-of-pictures sizes.

3. The computer-implemented method of claim 1 , wherein quantization values of the blocks on the identified point of interest are adjusted to be finer than quantization values of the blocks not on the identified point of interest.

4. The computer-implemented method of claim 1 , wherein numbers of P-frames of the blocks on the identified point of interest are adjusted to be higher than numbers of P-frames of the blocks not on the identified point of interest.

5. The computer-implemented method of claim 1 , wherein group-of-pictures sizes of the blocks on the identified point of interest are adjusted to be smaller than group-of-pictures sizes of the blocks not on the identified point of interest.

6. The computer-implemented method of claim 1 , wherein motion estimation of the blocks on the identified point of interest is adjusted to be finer than motion estimation of the blocks not on the identified point of interest.

7. The computer-implemented method of claim 1 , wherein the quality parameters on blocks not on the identified point of interest are adjusted to reduce a bit rate of the video data by an amount about equal to the amount that adjusting the quality parameters on blocks on the identified point of interest increased the bit rate of the video data.

8. The computer-implemented method of claim 1 , wherein the frames in the video data are additionally analyzed by an object segmenter to identify the point of interest in the frames of the specified type.

9. The computer-implemented method of claim 1 ,
wherein the specified type of point of interest is a type of subject in at least some frames of the video data,
wherein the type of subject in the at least some frames of the video data includes outdoors,
wherein the specified type of point of interest is based on a user profile,
wherein quantization values of the blocks on the identified point of interest are adjusted to be finer than quantization values of the blocks not on the identified point of interest.

10. A computer program product for compressing to compress video data, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, wherein the computer readable storage medium is not a transitory signal per se, the program instructions readable and/or executable by a computer to cause the computer to perform a method comprising:
receiving, by the computer, specification of a type of point of interest in video data,
wherein the specified type of point of interest is a type of subject in at least some frames of the video data,
wherein the type of subject in the at least some frames of the video data includes outdoors,

logically divide, by the computer, frames in the video data into a plurality of blocks;
analyzing, by the computer, frames in the video data to identify a point of interest in the frames of the specified type;
adjusting, by the computer, quality parameters on the blocks on the identified point of interest to improve a quality thereof,
wherein adjusting the quality parameters on the blocks on the identified point of interest to improve the quality thereof includes increasing a number of bits that represent a vector of the identified point of interest;
adjusting, by the computer, quality parameters on the blocks not on the identified point of interest to reduce a quality thereof; and
outputting, by the computer, a compressed video stream having the adjusted quality parameters.

11. The computer program product of claim 10 , wherein the quality parameters adjusted on blocks on the identified point of interest are selected from the group consisting of: quantization values of the blocks, number of P-frames, group-of-pictures sizes, and motion estimation.

12. The computer program product of claim 10 , wherein quantization values of the blocks on the identified point of interest are adjusted to be finer than quantization values of the blocks not on the identified point of interest.

13. The computer program product of claim 10 , wherein numbers of P-frames of the blocks on the identified point of interest are adjusted to be higher than numbers of P-frames of the blocks not on the identified point of interest.

14. The computer program product of claim 10 , wherein group-of-pictures sizes of the blocks on the identified point of interest are adjusted to be smaller than group-of-pictures sizes of the blocks not on the identified point of interest.

15. The computer program product of claim 10 ,
wherein motion estimation of the blocks on the identified point of interest is adjusted to be finer than motion estimation of the blocks not on the identified point of interest,
wherein the vector points in a first direction from a block of a first frame of the video data that indicates an estimated location of an associated block in a second frame of the video data,
wherein the first frame precedes the second frame.

16. The computer program product of claim 10 , wherein the quality parameters on blocks not on the identified point of interest are adjusted to reduce a bit rate of the video data by an amount about equal to the amount that adjusting the quality parameters on blocks on the identified point of interest increased the bit rate of the video data.

17. The computer program product of claim 10 , wherein the specified type of point of interest is based on a user profile.

18. The computer program product of claim 10 ,
wherein the frames in the video data are analyzed by an object detector to identify the point of interest in the frames of the specified type,
wherein the vector points in a first direction from a block of a first frame of the video data that indicates an estimated location of an associated block in a second frame of the video data,
wherein the first frame precedes the second frame,
wherein adjusting the quality parameters on blocks on the identified point of interest includes increasing an amount of motion estimation bits on blocks on the identified point of interest.

19. A system, comprising:
a processing circuit; and
logic integrated with the processing circuit, executable by the processing circuit, or integrated with and executable by the processing circuit, the logic being configured to cause the processing circuit to:
receive, by the processing circuit, specification of a type of point of interest in video data, wherein the specified type of poi

Assignees

• IBM

Inventors

• Liu Su
• Hwang Inseok
• Rozner Eric J
• Lee Jinho

Classifications

• H04N19/189 Primary

  characterised by the adaptation method, adaptation tool or adaptation type used for the adaptive coding · CPC title

• H04N19/513

  Processing of motion vectors · CPC title

• H04N19/176

  the region being a block, e.g. a macroblock · CPC title

• H04N19/114

  Adapting the group of pictures [GOP] structure, e.g. number of B-frames between two anchor frames (H04N19/107 takes precedence) · CPC title

• H04N19/167

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