| 摘要 |
<P>PROBLEM TO BE SOLVED: To provide an action data generation system capable of increasing greatly, compared with hitherto, the number of markers mountable on a subject, while keeping small the number of set colors of color markers. <P>SOLUTION: Each frame of animation data is divided into a plurality of blocks, and the image region of one marker is captured as a region spreading over the plurality of blocks B<SB>1</SB>-B<SB>4</SB>, and marker regions A<SB>1</SB>-A<SB>4</SB>in the blocks are extracted individually, and each position of the center of gravity G<SB>1</SB>-G<SB>4</SB>of each marker region A<SB>1</SB>-A<SB>4</SB>is determined. The positions derived from the same marker appear close on the image, and the positions derived from a different marker appear separately at a fixed distance. It can be discriminated based on each operation result G<SB>1</SB>-G<SB>4</SB>of the center of gravity for every marker region, whether the marker regions A<SB>1</SB>-A<SB>4</SB>in each block B<SB>1</SB>-B<SB>4</SB>are related with the same marker or not. The position of the center of gravity of a marker region not divided into blocks (namely, the center of gravity of the integrated area of the marker regions) can be simply operated by using each operation result G<SB>1</SB>-G<SB>4</SB>of the center of gravity of each discriminated marker region. <P>COPYRIGHT: (C)2004,JPO |
