Information processing apparatus, information processing method, information processing system, and program
AssigneeSony Corporation
InventorsTakayuki Ochi, Kazuhiro Nakagomi, Yorimitsu Naito, Takamasa Sato
Life of this patent
Estimating the term…
Abstract
There is provided an information processing apparatus including a face image acquisition section which acquires face images extracted from images which are classified into an identical time cluster by performing time clustering, and a person information generation section which classifies the face images for each time cluster, and generates person information in which persons regarded as an identical person are identified based on the face images which are classified.
Claims13
An information processing apparatus comprising:
- a central processing unit (CPU) operable to:
- classify a plurality of captured images into a plurality of identical time clusters, wherein an identical time cluster of the plurality of identical time clusters comprises a group of images which are clustered on the basis of date and/or time at which each of the images are captured;
- extract face regions from the group of images that belong to the identical time cluster of the plurality of identical time clusters;
- classify the extracted face regions into a plurality of face clusters, wherein a face cluster of the plurality of face clusters comprises face regions of an identical person;
- generate person information in which persons regarded as the identical person are identified based on the face regions of the identical person which are classified in the face cluster;
- and
- integrate a first face cluster of the plurality of face clusters with a second face cluster of the plurality of face clusters, which are not included in an identical image, based on average face feature quantities of the first face cluster and the second face cluster.
An information processing method comprising:
- classifying a plurality of captured images into a plurality of identical time clusters, wherein an identical time cluster of the plurality of identical time clusters comprises a group of images which are clustered on the basis of date and/or time at which each of the images are captured;
- extracting face regions from the group of images that belong to the identical time cluster of the plurality of identical time clusters;
- classifying the extracted face regions into a plurality of face clusters, wherein a face cluster of the plurality of face clusters comprises the face regions of an identical person;
- generating person information in which persons regarded as the identical person are identified based on the face regions of the identical person which are classified in the face cluster;
- and
- integrating a first face cluster of the plurality of face clusters with a second face cluster of the plurality of face clusters, which are not included in an identical image, based on average face feature quantities of the first face cluster and the second face cluster.
A non-transitory computer-readable storage medium, having stored thereon, a set of computer-executable instructions for causing a computer to perform a method comprising:
- classifying a plurality of captured images into a plurality of identical time clusters, wherein an identical time cluster of the plurality of identical time clusters comprises a group of images which are clustered on the basis of date and/or time at which each of the images are captured;
- extracting face regions from the group of images that belong to the identical time cluster of the plurality of identical time clusters;
- classifying the extracted face regions extracted from the identical time cluster into a plurality of face clusters, wherein a face cluster of the plurality of face clusters comprises the face regions of an identical person;
- generating person information in which persons regarded as the identical person are identified based on the face regions of the identical person which are classified in the face cluster;
- and
- integrating a first face cluster of the plurality of face clusters with a second face cluster of the plurality of face clusters, which are not included in an identical image, based on average face feature quantities of the first face cluster and the second face cluster.
Dependencies are inferred from the claim text.