摘要 |
<P>PROBLEM TO BE SOLVED: To generate a correct depth, based on a two-dimensional image. <P>SOLUTION: An image processing device 1 of the embodiment includes: an area detection part 12 for detecting a three-dimensional (3D) object area 103 of a 3D object included in an input image 100; a base depth imparting part 13 for imparting a base depth 151 being the basic structure of a depth to the input image 100; and a depth map generation part 14 for acquiring a 3D object depth value in the neighborhood 103c of a grounding position of the 3D object area 103, based on the base depth 151 imparted to the input image 100, setting the acquired 3D object depth value in the base depth 151 as a depth value Z of the 3D object area 103, and then, generating a depth map 110. <P>COPYRIGHT: (C)2012,JPO&INPIT |