摘要 |
PROBLEM TO BE SOLVED: To make an emphasized display, etc., without losing the naturalness of an image when an image is partially enlarged or reduced according to depth information by detecting two-dimensional shifts in video position between plural video frames and extracting the depth information from the detection information. SOLUTION: An image photographed by a camera, an image reproduced by a video reproducing device, etc., is inputted from a video input circuit 20 first. This image is stored in a frame memory 24. A correspondence point detecting circuit 26 and a motion detecting circuit 30 acquire the depth information on a subject from plural video frames read out of the frame memory 24. Then an image generating circuit 32 generates an image such as a viewpoint change image according to the depth information. In this case, viewpoint changing, enlargement, reduction, and segmentation are performed according to input from an instruction input part 34. This instruction is actualized through an existent user interface.
|