System and method for object matching

Patent metadata

Field	Value
Publication number	US-10043103-B2
Application number	US-201615189795-A
Country	US
Kind code	B2
Filing date	Jun 22, 2016
Priority date	Jun 26, 2015
Publication date	Aug 7, 2018
Grant date	Aug 7, 2018

Abstract

A computer implemented method of matching a tagged target group, selected from a plurality of gallery groups in a first camera field of view, with probe groups in a second camera field of view, by determining a group context for the tagged target group, determining, in the second camera field of view, probe contexts for the respective local probe groups in the second camera field of view; matching the group context against the probe contexts to produce respective similarity scores; and identifying based upon the similarity scores, the probe group that matches the tagged target group.

First claim

The invention claimed is:

1. A method of matching a tagged target group, selected from a plurality of gallery groups in a first camera field of view, with probe groups in a second camera field of view, the method comprising:
determining a group context for the tagged target group;
determining, in the second camera field of view, probe contexts for the respective probe groups in the second camera field of view;
matching, using one or more processors, the group context against the probe contexts to produce respective similarity scores;
identifying, based upon the similarity scores, the probe group that matches the tagged target group;
determining appearance models for the tagged target group and the probe groups;
matching the appearance model of the tagged target group against the appearance models of the probe groups to produce respective intra-group similarity distance scores; and
identifying, based upon the intra-group similarity scores, the probe group that matches the tagged target group.

2. The method according to claim 1 , wherein determining the group context further includes determining, in the second camera field of view, a predicted group context for the tagged target group.

3. The method according to claim 1 , wherein the group context and the probe context are dissimilarity contexts.

4. The method according to claim 1 , wherein the appearance model for the tagged target group is determined based upon the tagged target group in the first camera field of view.

5. The method according to claim 1 , wherein at least one of the predicted group context and the probe context are determined based upon filter criteria.

6. An apparatus for matching a tagged target group, selected from a plurality of gallery groups in a first camera field of view, with probe groups in a second camera field of view, the apparatus comprising:
one or more processors; and
memory storing instructions that, when executed by the one or more processors, cause the one or more processors to function as:
a first determining unit configured to determine a group context for the tagged target group,
a second determining unit configured to determine, in the second camera field of view, probe contexts for the respective probe groups in the second camera field of view,
a first matching unit configured to match the group context against the probe contexts to produce respective similarity scores,
a first identifying unit configured to identify, based upon the similarity scores, the probe group that matches the tagged target group,
a third determining unit configured to determine appearance models for the tagged target group and the probe groups,
a second matching unit configured to match the appearance model of the tagged target group against the appearance models of the probe groups to produce respective intra-group similarity distance scores, and
a second identifying unit configured to identify, based upon the intra-group similarity scores, the probe group that matches the tagged target group.

7. The apparatus according to claim 6 , wherein the first determining unit further is configured to determine, in the second camera field of view, a predicted group context for the tagged target group.

8. The apparatus according to claim 6 , wherein the group context and the probe context are dissimilarity contexts.

9. The apparatus according to claim 6 , wherein the appearance model for the tagged target group is determined based upon the tagged target group in the first camera field of view.

10. The apparatus according to claim 6 , wherein at least one of the predicted group context and the probe context are determined based upon filter criteria.

11. A non-transitory computer readable storage medium storing a computer executable program to perform a method of matching a tagged target group, selected from a plurality of gallery groups in a first camera field of view, with probe groups in a second camera field of view, the method comprising:
determining a group context for the tagged target group;
determining, in the second camera field of view, probe contexts for the respective probe groups in the second camera field of view;
matching, using one or more processors, the group context against the probe contexts to produce respective similarity scores;
identifying, based upon the similarity scores, the probe group that matches the tagged target group;
determining appearance models for the tagged target group and the probe groups;
matching the appearance model of the tagged target group against the appearance models of the probe groups to produce respective intra-group similarity distance scores; and
identifying, based upon the intra-group similarity scores, the probe group that matches the tagged target group.

Assignees

• Canon Kk

Inventors

• Dodballapur Veena Murthy Srinivasa
• Liao Kewen

Classifications

• G06V10/761

  Proximity, similarity or dissimilarity measures · CPC title

• G06V20/53 Primary

  Recognition of crowd images, e.g. recognition of crowd congestion · CPC title

• G06V10/751

  Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching · CPC title

• G06F18/22

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

• G06K9/6202 Primary

  Physics · mapped topic