| 摘要 |
The present invention discloses methods for generating a slide show of an image. With the methods, every still image could be displayed with the vivid effect of a dynamic video. The steps of one method according to the present invention comprise: performing a semantic analysis of a image for detecting semantic regions; selecting a first region and a second region from the semantic regions; determining a first and a second zooming levels of the first and second regions, respectively; and generating a slide show of the image by panning the image from the first region to the second region while zooming the image from the first zooming level to the second zooming level, gradually.
|
