Collation apparatus and method for the same, and image searching apparatus and method for the same
A collation apparatus that collates an input image with registered images includes: a feature extracting unit extracting feature quantities from a region of an object in the input image; a calculating unit calculating, per each of the feature quantities, similarity degrees between the feature quantities extracted from the input image and corresponding feature quantities of the registered image, for each of the registered images; an integrating unit integrating, using a parameter, the similarity degrees for the feature quantities; an estimation unit estimating as to whether or not the object in the input image is identical with an object in each of the registered images; and a parameter learning unit learning the parameters to be used by the integrating unit, based on an estimation result of the estimation unit and the similarity degree integrated by the integrating unit.
1. A collation apparatus for collating an input image with a plurality of registered images comprising:
a processor; and
a memory having stored thereon instructions that when executed by the processor, cause the processor to:
extract a plurality of feature quantities from a region of an object in the input image;
calculate, per each of the feature quantities, a plurality of similarity degrees between the plurality of feature quantities extracted from the input image and a corresponding plurality of feature quantities of a registered image among the plurality of registered images;
integrate, using a parameter, the plurality of similarity degrees corresponding to the plurality of feature quantities into an integrated similarity degree;
estimate as to whether the object in the input image is identical with an object in each of the plurality of registered images; and
learn the parameter to be used to integrate the plurality of similarity degrees, based on a result of the estimation by the estimation unit and the integrated similarity degree.
2. The apparatus according to claim 1 , further comprising instructions stored on the memory that when executed by the processor, cause the processor to extract an identifier of the object from a region in a predetermined positional relation to the region of the object in the input image, wherein
the processor estimates as to whether the object in the input image is identical with the object in each of the plurality of registered images, based on the extracted identifier and an identifier preliminary added to the object in the registered image.
3. The apparatus according to claim 2 , wherein
the object is a face of a person, and the processor extracts a player number, as the identifier, from a region of a body corresponding to the face.
4. The apparatus according to claim 1 , wherein
the processor estimates as to whether the object in the input image is identical with the object in each of the plurality of registered images, based on the integrated similarity degree.
5. The apparatus according to claim 1 , wherein the region of the object in the input image includes a plurality of local regions, and the extracted plurality of feature quantities include a feature quantity extracted from each of the plurality of local regions in the input image.
6. The apparatus according to claim 1 , wherein
the extracted plurality of feature quantities include a feature quantity relating to a property of the object.
7. The apparatus according to claim 1 , wherein
the input image is an image picked up by one of a plurality of imaging devices,
the memory has stored thereon instructions that when executed by the processor, cause the processor to acquire an identifier of the one imaging device that picked up the input image, and
the processor learns the parameter used to integrate the plurality of similarity degrees, per each acquired identifier.
8. An image searching apparatus configured to collate an input image with a plurality of registered images comprising:
a processor; and
a memory having stored thereon instructions that when executed by the processor, cause the processor to:
extract a plurality of feature quantities from a region of an object in the input image;
calculate, per each of the feature quantities, a plurality of similarity degrees between the plurality of feature quantities extracted from the input image and a corresponding plurality of feature quantities of a registered image among the plurality of registered images;
integrate, using a parameter, the plurality of similarity degrees corresponding to the plurality of feature quantities into an integrated similarity degree;
estimate as to whether the object in the input image is identical with an object in each of the plurality of registered images;
learn the parameter to be used to integrate the plurality of similarity degrees, based on a result of the estimation and the integrated similarity degree;
register the image; and
output one or more images arranged in a ranking, based on the integrated similarity degree between the input image and each of the plurality of registered images.
9. A collation method for collating an input image with a plurality of registered images comprising:
extracting a plurality of feature quantities from a region of an object in the input image;
calculating, per each of the feature quantities, a plurality of similarity degrees between the plurality of feature quantities extracted from the input image and a corresponding plurality of feature quantities of a registered image among the plurality of registered images;
integrating, using a parameter, the plurality of similarity degrees corresponding to the plurality of feature quantities into an integrated similarity degree;
estimating as to whether the object in the input image is identical with an object in each of the plurality of registered images; and
learning the parameter to be used in the integrating, based on a result of the estimation in the estimating and the integrated similarity degree.
10. An image searching method comprising:
extracting a plurality of feature quantities from a region of an object in the input image;
calculating, per each of the feature quantities, a plurality of similarity degrees between the plurality of feature quantities extracted from the input image and a corresponding plurality of feature quantities of a registered image among the plurality of registered images;
integrating, using a parameter, the plurality of similarity degrees corresponding to the plurality of feature quantities into an integrated similarity degree;
estimating as to whether the object in the input image is identical with an object in each of the plurality of registered images;
learning the parameter to be used in the integrating, based on a result of the estimation in the estimating and the integrated similarity degree;
registering the image; and
outputting one or more images arranged in a ranking, based on the integrated similarity degree between the input image and each of the plurality of registered images.
11. A non-transitory computer-readable recording medium storing a computer-readable program for causing a computer to:
extract a plurality of feature quantities from a region of an object in the input image;
calculate, per each of the feature quantities, a plurality of similarity degrees between the plurality of feature quantities extracted from the input image and a corresponding plurality of feature quantities of a registered image among the plurality of registered images;
integrate, using a parameter, the plurality of similarity degrees corresponding to the plurality of feature quantities into an integrated similarity degree;
estimate as to whether the object in the input image is identical with an object in each of the plurality of registered images; and
learn the parameter to be used to integrate the plurality of similarity degrees, based on a result of the estimation and the integrated similarity degree.
12. A non-transitory computer-readable recording medium storing a computer-readable program for causing a computer to:
extract a plurality of feature quantities from a region of an object in the input image;
calculate, per each of the feature quantities, a plurality of similarity degrees between the plurality of feature quantities extracted from the input image and a corresponding plurality of feature quantities of a registered image among the plurality of registered images;
integrate, using a parameter, the plurality of similarity degrees corresponding to the plurality of feature quantities into an integrated similarity degree;
estimate as to whether the object in the input image is identical with an object in each of the plurality of registered images;
learn the parameter to be used to integrate the plurality of similarity degrees, based on a result of the estimation and the integrated similarity degree;
register the image; and
output one or more images arranged in a ranking, based on the integrated similarity degree between the input image and each of the plurality of registered images.