Semantic Parsing of Objects in Video

• US 20160132730A1
• Filed: 01/18/2016
• Published: 05/12/2016
• Est. Priority Date: 07/28/2010
• Status: Active Grant

• Current Assignee: Kyndryl Incorporated
• Original Assignee: International Business Machines Corporation
• Inventors: Brown, Lisa Marie, Feris, Rogerio Schmidt, Hampapur, Arun, Vaquero, Daniel Andre
• Granted Patent: US 9,679,201 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 18/213 Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
• 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

Methods, systems, and computer program products for parsing objects are provided herein. A method includes producing a plurality of versions of an image of an object derived from an input, wherein each version comprises one of one multiple resolutions of said image of said object; computing an appearance probability at each of a plurality of regions on the one or more lowest resolution versions of said plurality of versions of said image for at least one attribute for said object; determining a configuration of the at least one attribute in the one or more lowest resolution versions based on at least the appearance probability in each of the plurality of regions; and outputting said configuration.

Claims

1. A method comprising:

producing a plurality of versions of an image of an object derived from an input, wherein each version comprises one of one multiple resolutions of said image of said object;

computing an appearance probability at each of a plurality of regions on the one or more lowest resolution versions of said plurality of versions of said image for at least one attribute for said object;

determining a configuration of the at least one attribute in the one or more lowest resolution versions based on at least the appearance probability in each of the plurality of regions; and

outputting said configuration.

2. The method of claim 1, comprising:

computing a geometric score for each region of said plurality of regions on the one or more lowest resolution versions, 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.

3. The method of claim 1, comprising:

analyzing one or more other versions of the multiple versions to compute a resolution context score for each of the plurality of regions in the one or more lowest resolution versions.

4. The method of claim 3, wherein the resolution context score for the one or more lowest resolution versions 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 attributes.

9. A computer program product comprising a computer readable storage hardware device having computer readable program code embodied in the computer readable storage hardware device, said computer readable program code containing instructions that perform a method for estimating parts and attributes of an object, said method comprising:

producing a plurality of versions of an image of an object derived from an input, wherein each version comprises one of one multiple resolutions of said image of said object;

computing an appearance probability at each of a plurality of regions on the one or more lowest resolution versions of said plurality of versions of said image for at least one attribute for said object;

determining a configuration of the at least one attribute in the one or more lowest resolution versions based on at least the appearance probability in each of the plurality of regions; and

outputting said configuration.

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

computing a geometric score for each region of said plurality of regions on the one or more lowest resolution versions, 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.

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

analyzing one or more other versions of the multiple versions to compute a resolution context score for each of the plurality of regions in the one or more lowest resolution versions.

12. The computer program product of claim 11, wherein the resolution context score for the one or more lowest resolution versions 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 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, said method comprising:

producing a plurality of versions of an image of an object derived from an input, wherein each version comprises one of one multiple resolutions of said image of said object;

computing an appearance probability at each of a plurality of regions on the one or more lowest resolution versions of said plurality of versions of said image for at least one attribute for said object;

determining a configuration of the at least one attribute in the one or more lowest resolution versions based on at least the appearance probability in each of the plurality of regions; and

outputting said configuration.

18. The computer system of claim 17, wherein said method further comprises:

analyzing one or more other versions of the multiple versions to compute a resolution context score for each of the plurality of regions in the one or more lowest resolution versions.