| 摘要 |
An image input unit acquires a first image and a second image. A matching processor computes a pixel matching between the images so that corresponding points are obtained on the second image with respect to lattice points of a mesh defined on the first image. A result thereof is recorded as a corresponding point file. An intermediate image generator generates an intermediate image of the first image and the second image based on the corresponding point file. Since the corresponding point file is composed of corresponding lattice points alone rather than corresponding pixels or the like, the amount of data is relatively small, and thereby allows efficient implementation of and application to various image effects, motion picture compression, on-line commodity presentation, and mobile applications.
|
