Collation apparatus, collation method, and computer program product

1. A collation apparatus comprising:

a user information storage medium configured to store user identification information relating to respective face feature vectors of faces of each of one or more users and corresponding object feature vectors; and

processing circuitry configured to at least:

receive a captured image including a face of a user and a non-facial object positioned in the image at a location different from a location of the face;

detect a face region from the image;

set an object region corresponding to the location of the non-facial object in the image;

extract a face feature vector from the face region and collate the extracted face feature vector with face feature vectors stored in the user information storage medium;

extract, from the object region corresponding to the location of the non-facial object, a plurality of images smaller than the object region, extract an object feature vector from each of the extracted images, and collate each extracted object feature vector with object feature vectors stored in the user information storage medium, wherein each object feature vector comprises an image-feature expressing a vector of the image capturing the non-facial object; and

calculate a collation result of the user using one or both of a collation result of face feature vectors and a collation result of object feature vectors.

2. The apparatus according to claim 1 , wherein the processing circuitry is further configured to display the image in which the object region is set, on a display.

3. The apparatus according to claim 1 , wherein the processing circuitry is further configured to determine the object region based on a position and a size of the face region and set the determined object region in the image.

4. The apparatus according to claim 1 , wherein the processing circuitry is further configured to collate each extracted object feature vector with object feature vectors for which the collation results correspond to a high-order predetermined number of face feature vectors among the object feature vectors stored in the user information storage medium.

5. The apparatus according to claim 1 , wherein the processing circuitry is further configured to collate each extracted object feature vector with object feature vectors for which the collation results correspond to the face feature vector equal to or greater than a threshold value among the object feature vectors stored in the user information storage medium.

6. The apparatus according to claim 1 , wherein the processing circuitry is further configured to

determine a plurality of object regions each corresponding to a location of a respective non-facial object and set the object regions in the image, and

extract object feature vectors from the respective object regions set in the image and collate each extracted object feature vector with the object feature vectors stored in the user information storage medium.

7. The apparatus according to claim 6 , wherein the processing circuitry is further configured to set the object region within the image among the object regions.

8. A collation method comprising:

receiving a captured image including a face of a user and a non-facial object positioned in the image at a location different from a location of the face;

detecting a face region from the image;

setting an object region corresponding to the location of the non-facial object in the image;

extracting a face feature vector from the face region;

collating the extracted face feature vector with face feature vectors stored in a user information storage medium that stores user identification information relating to respective face feature vectors of faces of each of one or more users and corresponding object feature vectors;

extracting, from the object region corresponding to the location of the non-facial object, a plurality of images smaller than the object region, and extracting an object feature vector from each of the extracted images, wherein each object feature vector comprises an image-feature expressing a vector of the image capturing the non-facial object;

collating each extracted object feature vector with object feature vectors stored in the user information storage medium; and

calculating a collation result of the user using one or both of a collation result of face feature vectors and a collation result of object feature vectors.

9. A computer program product comprising a non-transitory computer-readable storage medium containing a program which, when executed by a computer, causes the computer to execute at least:

receiving a captured image including a face of a user and a non-facial object positioned in the image at a location different from a location of the face;

detecting a face region from the image;

setting an object region corresponding to the location of the non-facial object in the image;

extracting a face feature vector from the face region;

collating the extracted face feature vector with face feature vectors stored in a user information storage medium that stores user identification information relating to respective face feature vectors of faces of one or more users and corresponding object feature vectors;

extracting, from the object region corresponding to the location of the non-facial object, a plurality of images smaller than the object region, and extracting an object feature vector from each of the extracted images, wherein each object feature vector comprises an image-feature expressing a vector of the image capturing the non-facial object;

collating each extracted object feature vector with object feature vectors stored in the user information storage medium; and

calculating a collation result of the user using one or both of a collation result of face feature vectors and a collation result of object feature vectors.