Method and apparatus for automatic visual artifact analysis and artifact reduction

Patent metadata

Field	Value
Publication number	US-9602838-B2
Application number	US-31264607-A
Country	US
Kind code	B2
Filing date	Dec 5, 2007
Priority date	Dec 28, 2006
Publication date	Mar 21, 2017
Grant date	Mar 21, 2017

Abstract

A system and method for correcting artifacts present in video frames is disclosed. The system includes a decision module ( 110 ) and an artifact correction unit ( 120 ). The system receives artifact metrics ( 112 ) and artifact maps ( 122 ) corresponding to artifacts present in video frames and prioritizes the video frames to be corrected by considering the type and severity of the artifacts and the content of the video frames ( 113 ). Subsequent to prioritizing the video frames, the system corrects artifacts of selected frames by adjusting compressed domain parameters ( 126 ), such as quantization parameters and mode decisions. An encoder ( 130 ) compresses the video frames according to the adjusted compressed domain parameter to automatically provide a video stream with a lower incidence and severity of artifacts.

First claim

The invention claimed is:

1. A video processing method comprising:
generating a plurality of metrics associated with a plurality of video compression artifact types present in a video picture included in a sequence of video pictures, wherein each metric indicates a severity of a respective one of said artifact types in said video picture and wherein each of said artifact types is characterized by a respective visible distortion of said artifact type; and
correcting said video picture in response to said plurality of metrics.

2. The method of claim 1 , wherein said artifact types include a first artifact type and a second artifact type, wherein said metrics include a first metric for said first artifact type and a second metric for said second artifact type, said method further comprising the steps of:
determining a content type categorizing said sequence of video pictures;
determining an average artifact level for said video picture in response to said first and second metrics, wherein said first metric for said first artifact type is weighed more than said second metric for said second artifact type in accordance with the content type of the sequence of video pictures when computing the average artifact level; and
selecting said video picture from the sequence of video pictures for correction in accordance with the determined average artifact level.

3. The method of claim 2 , wherein said selecting step is performed in response to prioritizing the pictures in accordance with said metrics.

4. The method of claim 2 , wherein said content type includes at least one of animation content type and movie content type.

5. The method of claim 1 , wherein said video compression artifact types comprise at least one of: a blocking artifact type, a banding artifact type, a dark patch artifact type and a ringing artifact type.

6. The method of claim 1 , wherein said correcting step includes utilizing an artifact map to correct a first block including an artifact of said picture.

7. The method of claim 6 , wherein said artifact map is segmented into blocks of a fixed size.

8. The method of claim 6 , wherein said correcting step further comprises correcting a second block that does not include an artifact.

9. The method of claim 1 , wherein said correcting step further comprises adjusting at least one of a quantization parameter and a mode decision corresponding to at least one block of said video picture.

10. The method of claim 1 , further comprising the steps of:
providing a user-interface; and
notifying a user of at least one type of artifact present in said video picture.

11. The method of claim 10 , including the step of:
classifying said video picture in accordance with the severity of at least one artifact; and
notifying the user of the classification of said video picture.

12. A processing system comprising:
a unit configured to generate compressed domain parameters in response to a plurality of metrics associated with a plurality of video compression artifact types present in a video picture included in a sequence of video pictures, wherein each metric indicates a severity of a respective one of said artifact types in said video picture and wherein each of said artifact types is characterized by a respective visible distortion of said artifact type; and
a processor configured to correct said video picture in accordance with said compressed domain parameters.

13. The system of claim 12 , wherein said artifact types include a first artifact type and a second artifact type, wherein said metrics include a first metric for said first artifact type and a second metric for said second artifact type, and wherein an artifact correction unit is configured to determine an average artifact level for said video picture in response to said first and second metrics, wherein said first metric for said first artifact type is weighed more than said second metric for said second artifact type in accordance with the content type of the sequence of video pictures when computing the average artifact level, and
select said video picture from the sequence of video pictures for correction in accordance with the determined average artifact level.

14. The system of claim 13 , wherein the artifact correction unit is configured to select said picture in response to prioritizing the pictures in accordance with said metrics.

15. The system of claim 13 , wherein said content type includes at least one of animation content type and movie content type.

16. The system of claim 12 , wherein said video compression artifact types include at least one of: a blocking artifact type, a banding artifact type, a dark patch artifact type and a ringing artifact type.

17. The system of claim 13 , wherein said artifact correction unit utilizes an artifact map to correct a first block including an artifact of said picture.

18. The system of claim 17 , wherein said artifact map is segmented into blocks of a fixed size.

19. The system of claim 17 , wherein said artifact correction unit is configured to correct a second block of said picture that does not include an artifact.

20. The system of claim 13 , wherein said artifact correction unit is configured to adjust at least one of a quantization parameter and a mode decision corresponding to at least one block of said video picture.

21. The system of claim 12 , comprising:
a user-interface configured to notify a user of at least one type of artifact present in said video picture.

22. The system of claim 21 , wherein said unit is configured to generate a classification of said video picture in accordance with the severity of at least one artifact and said user-interface is configured to notify the user of the classification of said video picture.

Assignees

• Dane Gokce
• Kumwilaisak Wuttipong
• Gomila Cristina
• Thomson Licensing

Inventors

• Dane Gokce
• Kumwilaisak Wuttipong
• Gomila Cristina

Classifications

• H04N19/154

  Measured or subjectively estimated visual quality after decoding, e.g. measurement of distortion (use of rate-distortion criteria H04N19/147) · CPC title

• H04N19/124

  Quantisation · CPC title

• H04N19/194

  involving only two passes · CPC title

• H04N19/61

  in combination with predictive coding · CPC title

• H04N19/86 Primary

  involving reduction of coding artifacts, e.g. of blockiness · CPC title