Static image segmentation

Patent metadata

Field	Value
Publication number	US-9311716-B2
Application number	US-201414277321-A
Country	US
Kind code	B2
Filing date	May 14, 2014
Priority date	May 14, 2014
Publication date	Apr 12, 2016
Grant date	Apr 12, 2016

Abstract

Methods, systems, and computer program products for static image segmentation are provided herein. A method includes segmenting a static image containing a target object into multiple regions based on one or more visual features of the static image; analyzing video content containing the target object to determine a similarity metric across the multiple segmented regions based on motion information associated with each of the multiple segmented regions; and applying the similarity metric to the static image to identify two or more of the multiple segmented regions as being portions of the target object.

First claim

What is claimed is:

1. A method comprising:
segmenting a static image containing a target object into multiple regions based on one or more visual features of the static image;
analyzing video content containing the target object to determine a similarity metric across the multiple segmented regions based on motion information associated with each of the multiple segmented regions; and
applying the similarity metric to the static image to identify two or more of the multiple segmented regions as being portions of the target object;
wherein at least one of said segmenting, said analyzing, and said applying is carried out by a computing device.

2. The method of claim 1 , wherein said segmenting based on one or more visual features comprises identifying multiple edges within the static image.

3. The method of claim 1 , wherein said segmenting based on one or more visual features comprises identifying multiple colors within the static image.

4. The method of claim 1 , wherein said segmenting based on one or more visual features comprises identifying multiple shapes within the static image.

5. The method of claim 1 , wherein segmenting based on said one or more visual features comprises identifying multiple textures within the static image.

6. The method of claim 1 , wherein said segmenting comprises applying a hierarchical segmentation technique to the static image.

7. The method of claim 1 , wherein said segmenting comprises applying a superpixel segmentation technique to the static image.

8. The method of claim 1 , wherein said determining the similarity metric comprises determining a level of consistency in the motion information in two or more of the multiple segmented regions.

9. The method of claim 8 , comprising:
comparing said level of consistency to a predetermined threshold value.

10. The method of claim 1 , wherein said motion information associated with each of the multiple segmented regions comprises one or more statistical patterns associated with each of the multiple segmented regions.

11. The method of claim 1 , wherein said determining the similarity metric comprises applying a feature-based clustering technique to the video content.

12. The method of claim 1 , wherein said determining the similarity metric comprises applying a motion displacement-based clustering technique to the video content.

13. The method of claim 1 , comprising:
searching for video content containing the target object.

14. The method of claim 1 , comprising:
identifying the multiple segmented regions in the video content.

15. A computer program product, the computer program product comprising a non-transitory computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computing device to cause the computing device to:
segment a static image containing a target object into multiple regions based on one or more visual features of the static image;
analyze video content containing the target object to determine a similarity metric across the multiple segmented regions based on motion information associated with each of the multiple segmented regions; and
apply the similarity metric to the static image to identify two or more of the multiple segmented regions as being portions of the target object.

16. The computer program product of claim 15 , wherein said segmenting comprises applying a superpixel segmentation technique to the static image.

17. The computer program product of claim 15 , wherein said determining the similarity metric comprises determining a level of consistency in the motion information in two or more of the multiple segmented regions.

18. The computer program product of claim 15 , wherein said determining the similarity metric comprises applying a feature-based clustering technique to the video content.

19. The computer program product of claim 15 , wherein said determining the similarity metric comprises applying a motion displacement-based clustering technique to the video content.

20. A system comprising:
a memory; and
at least one processor coupled to the memory and configured for:
segmenting a static image containing a target object into multiple regions based on one or more visual features of the static image;
analyzing video content containing the target object to determine a similarity metric across the multiple segmented regions based on motion information associated with each of the multiple segmented regions; and
applying the similarity metric to the static image to identify two or more of the multiple segmented regions as being portions of the target object.

Assignees

• IBM

Inventors

• Codella Noel C
• Hua Gang
• Smith John R

Classifications

• G06F18/23

  Clustering techniques · CPC title

• G06F18/22

  Matching criteria, e.g. proximity measures · CPC title

• G06V10/255

  Detecting or recognising potential candidate objects based on visual cues, e.g. shapes · CPC title

• G06V10/422

  for representing the structure of the pattern or shape of an object therefor · CPC title

• G06V10/759

  Region-based matching · CPC title