JP2015204030A5

JP2015204030A5 -

Info

Publication number: JP2015204030A5
Application number: JP2014083781A
Authority: JP
Filing date: 2014-04-15
Publication date: 2017-05-25
Anticipated expiration: 2034-04-15

Claims

Candidate area extracting means for extracting a candidate area of a subject from an image;
Feature quantity extraction means for extracting a feature quantity relating to the attribute of the image from the candidate area of the subject extracted by the candidate area extraction means;
Attribute determination means for determining attributes of the candidate area of the subject extracted by the candidate area extraction means based on the feature quantity extracted by the feature quantity extraction means;
A recognition apparatus comprising: determination result integration means for identifying image attributes by integrating the determination results of the attribute determination means.

The recognition apparatus according to claim 1, wherein the candidate area extraction unit extracts candidate areas of a plurality of subjects based on a criterion of a predetermined candidate area.

The recognition apparatus according to claim 2, wherein the candidate area extracting unit includes a plurality of candidate area extracting units based on a plurality of different criteria.

The recognition apparatus according to claim 3, wherein the attribute determination unit includes a plurality of attribute determination units respectively corresponding to a plurality of different candidate area extraction units.

The candidate area extraction unit extracts the plurality of candidate areas by repeating a process of combining two adjacent superpixels having the highest similarity among the plurality of superpixels generated by dividing the image. The recognition apparatus according to claim 2.

The recognition apparatus according to claim 5, wherein the candidate area extracting unit extracts, as the plurality of candidate areas, a super pixel that is generated by repeating the processing and has a larger area than a predetermined value among the super pixels. .

The candidate region extraction means, recognition apparatus according to any one of claim 1 to 6, characterized in that it comprises a body region extracting means for extracting a human candidate region from the image.

Said attribute determining means, any claim 1-7, characterized in that for performing the learning by the candidate region when learning the attribute of the candidate region is classified as divided according to the attribute of the image The recognition apparatus of Claim 1.

The recognition apparatus according to claim 8 , wherein the attribute determination unit performs learning so as to preferentially classify a region taught as an important candidate region when classifying the candidate region.

It said attribute determining means, recognition apparatus according to any one of claim 1 to 9, characterized in that learning is performed based on the candidate region of an object not associated with the attribute of the image.

It said attribute determining means, recognition apparatus according to any one of claim 1 to 10, characterized in that it is constituted by a method based on classification tree.

It said attribute determining means, recognition apparatus according to any one of claim 1 to 10, characterized in that it is constituted by a method based on the search of the similar case data.

It said attribute determining means, recognition apparatus according to any one of claim 1 to 10, characterized in that it is constituted by a method based on hashing.

The attribute to be determined by the attribute determining means is any one of an image scene, a crowd action in the image, a composition type of the image, information on a main subject of the image, and information on a light source direction of the image The recognition device according to any one of claims 1 to 13 .

Recognition apparatus according to any one of claim 1 to 14, wherein the target image determining said attribute is a motion picture.

Recognition apparatus according to any one of claim 1 to 14, wherein the target image determining said attribute is a range image.

A candidate area extracting step of extracting a candidate area of the subject from the image;
A feature amount extraction step of extracting a feature amount related to the attribute of the image from the candidate region of the subject extracted in the candidate region extraction step;
An attribute determination step of determining attributes of the candidate region of the subject extracted in the candidate region extraction step based on the feature amount extracted in the feature amount extraction step;
A recognition method comprising: a determination result integration step for identifying image attributes by integrating the determination results of the attribute determination step.

A candidate area extracting step of extracting a candidate area of the subject from the image;
A feature amount extraction step of extracting a feature amount related to the attribute of the image from the candidate region of the subject extracted in the candidate region extraction step;
An attribute determination step of determining attributes of the candidate region of the subject extracted in the candidate region extraction step based on the feature amount extracted in the feature amount extraction step;
A program for causing a computer to execute a determination result integration step for identifying image attributes by integrating the determination results of the attribute determination step.