| 摘要 |
<P>PROBLEM TO BE SOLVED: To provide a composite image creating apparatus capable of readily converting video content into read contents that are easy to understand on the part of the users. <P>SOLUTION: A control section 112 acquires a caption and display timing of the caption from video content, captures a still image from the video contents, on the basis of display timing of the acquired caption, and makes the captured caption associate with the captured still image. When a plurality of still images captured from the video contents are similar to one another, a composited image is generated, by combining one of the plurality of still images with a caption associated with each of the plurality of still images. When whether the plurality of still images captured from the video contents are similar to one another is determined, threshold which is used for the similarity determination is changed dynamically, according to the kind of video content. <P>COPYRIGHT: (C)2010,JPO&INPIT |
