| 摘要 |
<P>PROBLEM TO BE SOLVED: To easily apply annotations to contents and provide applications using the annotations. <P>SOLUTION: A learning device 312 extracts an image feature amount of each frame of images of learning contents and also extracts word frequency information which relates to the frequency of appearance of each word in explanation texts for explaining the details of the images of the learning contents as a text feature amount of the explanation texts and learns an annotation model which is a multi-stream HMM by using the multi-stream having the image feature amount and text feature amount. A browsing control device 314 extracts a scene which is a collection of more than one temporally continuous frames from a target content by using the annotation model and displays representative images of the scene in time order. The present invention is applicable, for example, to the case in which annotations are applied to contents. <P>COPYRIGHT: (C)2012,JPO&INPIT |
