Video quality measurement

Patent metadata

Field	Value
Publication number	US-10075710-B2
Application number	US-201114353351-A
Country	US
Kind code	B2
Filing date	Nov 24, 2011
Priority date	Nov 24, 2011
Publication date	Sep 11, 2018
Grant date	Sep 11, 2018

Abstract

A particular implementation receives a bitstream and derives parameters from the bitstream. The parameters include quantization parameters, content unpredictability parameters, ratios of lost blocks, ratios of propagated blocks, error concealment distances, motion vectors, durations of freezing, and frame rates. Using these parameters, a compression distortion factor, a slicing distortion factor, and a freezing distortion factor are estimated respectively for distortions resulting from video compression, a slicing mode error concealment, and a freezing slicing mode error concealment. The distortion factors are then mapped to a composite video quality score. For applications with limited computational power, the estimation of distortion factors can be simplified. In particular, the compression distortion factor, the slicing distortion factor, and the freezing distortion factor can be predicted from quantization parameters, ratios of lost blocks, and durations of freezing, respectively.

First claim

The invention claimed is:

1. A method, executed by one or more video processors, comprising:
determining parameters for a video bitstream without reconstructing a video corresponding to the video bitstream, the parameters including durations of freezing, motion vectors in pictures immediately before freezing, ratios of propagated blocks, and a frame rate, wherein propagated blocks are blocks that directly or indirectly use lost blocks for prediction;
determining a slicing distortion factor and a freezing distortion factor for the video bitstream without reconstructing a video corresponding to the video bitstream, the slicing distortion factor representative of distortion resulting from a slicing mode error concealment and the freezing distortion factor representative of distortion resulting from a freezing mode error concealment, wherein the freezing distortion factor is determined in response to the durations of freezing, the motion vectors, and the frame rate, and wherein the slicing distortion factor is determined in response to the ratios of propagated blocks; and
determining a quality metric for the video in response to the slicing distortion factor and the freezing distortion factor.

2. The method of claim 1 , wherein the determined parameters further include ratios of lost blocks, and wherein the slicing distortion factor is determined further in response to the ratios of lost blocks.

3. The method of claim 2 , wherein the slicing distortion factor is determined in response to a power function of the ratios of lost blocks and the freezing distortion factor is determined in response to a power function of the durations of freezing.

4. The method of claim 2 , the determined parameters further comprising error concealment distances, and content unpredictability parameters, wherein the slicing distortion factor is determined in response to ones of the ratios of lost blocks, the ratios of propagated blocks, the error concealment distances, the content unpredictability parameters, and the frame rate.

5. The method of claim 4 , further comprising:
decoding a plurality of pictures from the video bitstream;
detecting mosaic artifacts in the plurality of pictures; and
updating at least one of the ratios of lost blocks and the ratios of propagated blocks based on the detecting.

6. The method of claim 1 , wherein the determining the quality metric is further in response to a compression distortion factor representative of distortion resulting from compression.

7. The method of claim 6 , wherein the quality metric is determined in response to a weighted sum of the compression distortion factor, the slicing distortion factor, and the freezing distortion factor.

8. The method of claim 7 , wherein the quality metric is determined in response to γ/(1+αd β ), wherein d corresponds to the weighted sum and α, β, and γ are constants.

9. The method of claim 1 , further comprising:
performing at least one of monitoring quality of the video bitstream, adjusting the video bitstream in response to the quality metric, creating a new video bitstream based on the quality metric, adjusting parameters of a distribution network used to transmit the video bitstream, determining whether to keep the video bitstream based on the quality metric, and choosing an error concealment mode at a decoder.

10. An apparatus, comprising at least one memory and one or more video processors, the one or more video processors being configured to:
determine parameters for a video bitstream without reconstructing a video corresponding to the video bitstream, the parameters including durations of freezing, motion vectors in pictures immediately before freezing, ratios of propagated blocks, and a frame rate, wherein propagated blocks are blocks that directly or indirectly use lost blocks for prediction;
determine, without reconstructing a video corresponding to the video bitstream, a slicing distortion factor representative of distortion resulting from a slicing mode error concealment;
determine, without reconstructing the video corresponding to the video bitstream, a freezing distortion factor representative of distortion resulting from a freezing mode error concealment, wherein the freezing distortion factor is determined in response to the durations of freezing, the motion vectors, and the frame rate, and wherein the slicing distortion factor is determined in response to the ratios of propagated blocks; and
determine a quality metric for said video in response to the slicing distortion factor and the freezing distortion factor.

11. The apparatus of claim 10 , wherein the determined parameters further include ratios of lost blocks, and wherein the slicing distortion factor is determined further in response to the ratios of lost blocks.

12. The apparatus of claim 11 , the processor further determining at least one of error concealment distances, and content unpredictability parameters, wherein the slicing distortion factor is determined in response to ones of the ratios of lost blocks, the ratios of propagated blocks, the error concealment distances, the frame rate, and the content unpredictability parameters.

13. The apparatus of claim 11 , wherein the one or more video processors are further configured to:
decode a plurality of pictures from the video bitstream; and
detect mosaic artifacts in the plurality of pictures, wherein the ratios of lost blocks and the ratios of propagated blocks are updated based on the detecting.

14. The apparatus of claim 10 , wherein a compression distortion factor representative of distortion resulting from compression is determined.

15. The apparatus of claim 14 , wherein the quality metric is determined in response to γ/(1+αd β ), where d corresponds to a weighted sum of the slicing distortion factor, the freezing distortion factor, and the compression distortion factor.

16. The apparatus of claim 10 , wherein the one or more video processors are further configured to perform at least one of monitoring quality of the video bitstream, adjusting the video bitstream in response to the quality metric, creating a new video bitstream based on the quality metric, adjusting parameters of a distribution network used to transmit the video bitstream, determining whether to keep the video bitstream based on the quality metric, and choosing an error concealment mode at a decoder.

17. A non-transitory processor readable medium having stored thereupon instructions for causing one or more processors to collectively perform:
determining parameters for a video bitstream without reconstructing a video corresponding to the video bitstream, the parameters including durations of freezing, motion vectors in pictures immediately before freezing, ratios of propagated blocks, and a frame rate, wherein propagated blocks are blocks that directly or indirectly use lost blocks for prediction;
determining at least one of a slicing distortion factor and a freezing distortion factor for the video bitstream without reconstructing a video corresponding to the video bitstream, the slicing distortion factor representative of distortion resulting from a slicing mode error concealment and the freezing distortion factor representative of distortion resulting from a freezing mode error concealment, wherein the freezing distortion factor is determined in response to the durations of freezing, the motion vectors, and the frame rate, and wherein the slicing distortion factor is determined in response to the ratios of propagated blocks; and
determining a quality metric for the video in response to the slicing distortion factor an

Assignees

• Zhang Fan
• Liao Ning
• Xie Kai
• Chen Zhibo
• Thomson Licensing

Inventors

• Zhang Fan
• Liao Ning
• Xie Kai
• Chen Zhibo

Classifications

• H04N17/004 Primary

  for digital television systems · CPC title

• H04N19/00933 Primary

  Electricity · mapped topic

• H04N19/895

  in combination with error concealment · CPC title

• H04N19/176 Primary

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