| 摘要 |
To make it possible to accurately track a presented part, a low-resolution overall image is taken in from an input image, and a medium-resolution image, whose center is the head part, is extracted from it. When it is possible to extract a medium-resolution image in the middle of which is the head part, an even higher-resolution image is extracted, in the middle of which is the two-eyes part. <IMAGE> |
