Key-frame extraction from video

• US 20070263128A1
• Filed: 05/12/2006
• Published: 11/15/2007
• Est. Priority Date: 05/12/2006
• Status: Active Grant

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Zhang, Tong
• Granted Patent: US 8,379,154 B2
• Time in Patent Office: Days
• US Class Current: 348/700
• CPC Class Codes: G11B 27/28 by using information signals recorded by the same method as the main recording G11B27/22 takes precedence

Abstract

Techniques that enable flexible key-frame extraction from video. Key-frame extraction according to the present teachings includes segmenting a video into a set of shots each having a set of video frames and selecting a set of key-frames from the shots in response to an image quality of the corresponding video frames. The present techniques include a set of user selectable modes that enable key-frame extraction to be adapted to an underlying platform.

Claims

1. A method for key-frame extraction, comprising:

segmenting a video into a set of shots each having a set of video frames;

selecting a set of key-frames from the shots in response to an image quality of the corresponding video frames.

2. The method of claim 1, wherein selecting includes selecting the key-frames in response to a video content change in the corresponding video frames.

3. The method of claim 2, wherein selecting includes determining the video content change by determining a histogram difference in the corresponding video frames.

4. The method of claim 1, wherein selecting includes selecting the key-frames in response to an entropy for each corresponding video frame.

5. The method of claim 1, wherein selecting includes selecting the key-frames in response to a sharpness for each corresponding video frame.

6. The method of claim 1, wherein selecting includes selecting the key-frames in response to a contrast for each corresponding video frame.

7. The method of claim 1, wherein selecting includes selecting the key-frames in response to a user selected mode.

8. The method of claim 7, wherein selecting a set of key-frames includes selecting a fixed number of key-frames for the video as specified by the user selected mode.

9. The method of claim 7, wherein selecting includes selecting a fixed rate of key-frames for the video as specified by the user selected mode.

10. The method of claim 7, wherein selecting includes selecting one key-frame for each shot in the video as specified by the user selected mode.

11. A video system including a key-frame extractor that segments a video into a set of shots each having a set of video frames and that selects a set of key-frames from the shots in response to an image quality of the corresponding video frames.

12. The video system of claim 11, wherein the key-frame extractor selects the key-frames in response to a video content change in the corresponding video frames.

13. The video system of claim 12, wherein the key-frame extractor selects the key-frames in response to a user selected mode.

14. The video system of claim 13, wherein the user selected mode specifies a fixed number of key-frames for the video.

15. The video system of claim 14, wherein the key-frame extractor selects the key-frames by ranking the shots according length and content.

16. The video system of claim 15, wherein the key-frame extractor selects the key-frames by splitting one of the shots into two shots at a video frame having a greatest content change.

17. The video system of claim 14, wherein the key-frame extractor selects the key-frames by removing a key-frame having a low image quality.

18. The video system of claim 14, wherein the key-frame extractor selects the key-frames by removing a key-frame that corresponds to a short shot.

19. The video system of claim 13, wherein the user selected mode specifies a fixed rate of key-frames for the video.

20. The video system of claim 19, wherein the key-frame extractor selects the key-frames by dividing the video into a set of segments having a fixed length in response to the fixed rate by allocating one key-frame to each of a set of segments in the video in response to a set of shot boundaries and content changes.