Semantic Parsing of Objects in Video

• US 20130177249A1
• Filed: 03/04/2013
• Published: 07/11/2013
• Est. Priority Date: 07/28/2010
• Status: Active Grant

• Current Assignee: Kyndryl Incorporated (Kyndryl Holdings, Inc.)
• Original Assignee: International Business Machines Corporation
• Inventors: Brown, Lisa Marie, Feris, Rogerio Schmidt, Hampapur, Arun, Vaquero, Daniel Andre
• Granted Patent: US 8,588,533 B2
• Time in Patent Office: Days
• US Class Current: 382/195
• CPC Class Codes: G06F 18/22 Matching criteria, e.g. proximity measures
• CPC Class Codes: G06V 10/426 Graphical representations
• CPC Class Codes: G06V 20/10 Terrestrial scenes scenes under surveillance with static cameras G06V20/52; scenes perceived from the exterior of a vehicle G06V20/56; scenes perceived from the interior of a vehicle G06V20/59
• CPC Class Codes: G06V 20/41 Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items segmenting video sequences G06V20/49
• CPC Class Codes: G06V 20/70 Labelling scene content, e.g. deriving syntactic or semantic representations
• CPC Class Codes: G06V 30/2504 Coarse or fine approaches, e.g. resolution of ambiguities or multiscale approaches
• CPC Class Codes: G06V 40/103 Static body considered as a whole, e.g. static pedestrian or occupant recognition

Abstract

Techniques, systems, and computer program products for parsing objects in a video are provided herein. A method includes producing and storing a plurality of versions of an image of an object derived from a video input, wherein each version of said image has a different resolution of said image; computing an appearance score at each of a plurality of regions on the lowest resolution version of said image for a plurality of semantic attributes with associated parts for said object, said appearance score denoting a probability of each semantic attribute appearing in the region; analyzing increasingly higher resolution versions than the lowest resolution version to compute a resolution context score for each region in the lowest resolution version; and ascertaining an optimized configuration of body parts and associated semantic attributes in the lowest resolution version, said ascertaining utilizing the appearance scores and the resolution context scores.

Claims

1. A method comprising:

producing and storing a plurality of versions of an image of an object derived from a video input, said image cropped from said video input, and wherein each version of said image has a different resolution of said image of said object;

computing an appearance score at each of a plurality of regions on the lowest resolution version of said versions of said image for a plurality of semantic attributes with associated parts for said object, said appearance score for at least one semantic attribute of the plurality of semantic attributes for each region denoting a probability of each semantic attribute of the at least one semantic attribute appearing in the region;

analyzing increasingly higher resolution versions than the lowest resolution version to compute a resolution context score for each region in the lowest resolution version, said resolution context score being indicative of an extent to which finer spatial structure exists in the increasingly higher resolution versions than in the lowest resolution version for each region; and

ascertaining an optimized configuration of body parts and associated semantic attributes in the lowest resolution version, said ascertaining utilizing the appearance scores and the resolution context scores in the regions in the lowest resolution version.

2. The method of claim 1, comprising:

displaying and/or storing said optimized configuration of body parts and associated semantic attributes.

3. The method of claim 1, comprising:

computing a geometric score for each region of said plurality of regions on the lowest resolution version, said geometric score computing a probability of a region matching stored reference data for a reference object corresponding to the detected object with respect to angles and distances among the plurality of regions.

4. The method of claim 3, wherein the resolution context score for the lower resolution version of said image is computed as a weighted average score computed from a plurality of scores for a next higher resolution version of said higher resolution versions of said image.

5. The method of claim 4, wherein said plurality of scores for said next higher resolution version of said image comprise appearance scores and geometric scores.

6. The method of claim 4, wherein said plurality of scores for said next higher resolution version of said image comprise appearance scores, geometric scores and resolution context scores.

7. The method of claim 6, wherein said weighted average score for the next higher resolution version of the image is computed using the following formula divided by I:

8. The method claim of 7, comprising:

storing and/or displaying output of at least one portion of said image in at least one version of said higher level versions of said image with spatial information on semantic attributes and associated parts.

9. A computer program product comprising:

a computer readable storage medium having computer readable program code embodied in the storage medium, said computer readable program code containing instructions that perform a method for estimating parts and attributes of an object in video, said method comprising;

producing and storing a plurality of versions of an image of an object derived from a video input, said image cropped from said video input, and wherein each version of said image has a different resolution of said image of said object;

computing an appearance score at each of a plurality of regions on the lowest resolution version of said versions of said image for a plurality of semantic attributes with associated parts for said object, said appearance score for at least one semantic attribute of the plurality of semantic attributes for each region denoting a probability of each semantic attribute of the at least one semantic attribute appearing in the region;

analyzing increasingly higher resolution versions than the lowest resolution version to compute a resolution context score for each region in the lowest resolution version, said resolution context score being indicative of an extent to which finer spatial structure exists in the increasingly higher resolution versions than in the lowest resolution version for each region; and

ascertaining an optimized configuration of body parts and associated semantic attributes in the lowest resolution version, said ascertaining utilizing the appearance scores and the resolution context scores in the regions in the lowest resolution version.

10. The computer program product of claim 9, said method comprising:

displaying and/or storing said optimized configuration of body parts and associated semantic attributes.

11. The computer program product of claim 9, said method comprising:

computing a geometric score for each region of said plurality of regions on the lowest resolution version, said geometric score computing a probability of a region matching stored reference data for a reference object corresponding to the detected object with respect to angles and distances among the plurality of regions.

12. The computer program product of claim 11, wherein the resolution context score for the lower resolution version of said image is computed as a weighted average score computed from a plurality of scores for a next higher resolution version of said higher resolution versions of said image.

13. The computer program product of claim 12, wherein said plurality of scores for said next higher resolution version of said image comprise appearance scores and geometric scores.

14. The computer program product of claim 12, wherein said plurality of scores for said next higher resolution version of said image comprise appearance scores, geometric scores and resolution context scores.

15. The computer program product of claim 14, wherein said weighted average score for the next higher resolution version of the image is computed using the following formula divided by I:

16. The computer program product of claim 15, said method comprising:

storing and/or displaying output of at least one portion of said image in at least one version of said higher level versions of said image with spatial information on semantic attributes and associated parts.

17. A computer system comprising a processor and a computer readable memory unit coupled to the processor, said computer readable memory unit containing instructions that when run by the processor implement a method for estimating parts and attributes of an object in video, said method comprising:

producing and storing a plurality of versions of an image of an object derived from a video input, said image cropped from said video input, and wherein each version of said image has a different resolution of said image of said object;

computing an appearance score at each of a plurality of regions on the lowest resolution version of said versions of said image for a plurality of semantic attributes with associated parts for said object, said appearance score for at least one semantic attribute of the plurality of semantic attributes for each region denoting a probability of each semantic attribute of the at least one semantic attribute appearing in the region;

analyzing increasingly higher resolution versions than the lowest resolution version to compute a resolution context score for each region in the lowest resolution version, said resolution context score being indicative of an extent to which finer spatial structure exists in the increasingly higher resolution versions than in the lowest resolution version for each region; and

ascertaining an optimized configuration of body parts and associated semantic attributes in the lowest resolution version, said ascertaining utilizing the appearance scores and the resolution context scores in the regions in the lowest resolution version.

18. The system of claim 17, said method comprising:

displaying and/or storing said optimized configuration of body parts and associated semantic attributes.

19. A process for supporting computer infrastructure, said process comprising providing at least one support service for at lease one of creating, integrating, hosting, maintaining, and deploying computer-readable code in a computer system, wherein the code in combination with the computing system is capable of performing a method for estimating parts and attributes of an object in video, said method comprising:

producing and storing a plurality of versions of an image of an object derived from a video input, said image cropped from said video input, and wherein each version of said image has a different resolution of said image of said object;

computing an appearance score at each of a plurality of regions on the lowest resolution version of said versions of said image for a plurality of semantic attributes with associated parts for said object, said appearance score for at least one semantic attribute of the plurality of semantic attributes for each region denoting a probability of each semantic attribute of the at least one semantic attribute appearing in the region;

analyzing increasingly higher resolution versions than the lowest resolution version to compute a resolution context score for each region in the lowest resolution version, said resolution context score being indicative of an extent to which finer spatial structure exists in the increasingly higher resolution versions than in the lowest resolution version for each region; and

ascertaining an optimized configuration of body parts and associated semantic attributes in the lowest resolution version, said ascertaining utilizing the appearance scores and the resolution context scores in the regions in the lowest resolution version.

20. The process of claim 19, said method comprising:

displaying and/or storing said optimized configuration of body parts and associated semantic attributes.