Loading patent…
AssigneeSony Corporation
InventorsKazumasa Tanaka, Kenji Tanaka, Yoshihiro Takahashi
Estimating the term…
The present invention relates to an apparatus and a method of content retrieval and a program which facilitate recognizing content details of a moving image by displaying a retrieval result for each of a plurality of images. A metadata extracting unit 22 extracts metadata such as information of a representative image from content which is supplied from a content input unit 21 and stores the metadata in a metadata database 24. An image feature quantity extracting unit 23 extracts image feature quantity from content and stores the image feature quantity in an image feature quantity database 25 while extracting image feature quantity from a retrieval image as well. A retrieving unit 27 compares the image feature quantity of the retrieval image with the image feature quantity stored in the image feature quantity database 25 and specifies a still image which is similar to the retrieval image. An output unit 28 reads metadata of the content, generates display data to be necessary for performing timeline displaying, and further, generates display data for displaying a retrieval result on a timeline.
A content retrieval apparatus, comprising:
A content retrieval method, comprising the steps of:
A program for causing a computer to function as a content retrieval apparatus, the apparatus comprising:
Dependencies are inferred from the claim text.