| 摘要 |
<p><P>PROBLEM TO BE SOLVED: To provide a method for preventing, without complicating computation, mismatching of a predictive image caused by a computational precision difference of a motion vector in a motion compensation system that there is no limitation that all pixels belonging to the same patch have a common motion vector, and a horizontal/vertical component of the motion vector of a pixel may take a value other than an integer multiple of a distance between adjacent pixels. <P>SOLUTION: The horizontal and vertical components of the motion vector of each pixel used to synthesize the predictive image are specified to take only the values of integer multiples of 1/d1, 1/d2 (d1, d2 are a positive integer) of the distances between adjacent pixels. Mismatching of the predictive image caused by the computational precision difference of the motion vector can be prevented by simple computation, thereby simplifying processing as a result. <P>COPYRIGHT: (C)2008,JPO&INPIT</p> |
