| 摘要 |
An encoding/decoding apparatus is described wherein an encoding apparatus comprises a feature-image-extracting section, a three-dimensional shape model adjustment section, a brightness information detection section, a patch dividing section and a threshold processing section, and a decoding apparatus comprises a three-dimensional shape model forming section and a synthesizing section. Feature image information extracted from input signals by the extracting section is adjusted with three-dimensional shape basic model information by the adjustment section and then a brightness diffusion value for each triangle patch constituting the adjusted three-dimensional shape model is detected by the detection section. When the detected diffusion value is larger than a predetermined value, the processing section subjects the dividing section to divide the patch into three small patches, and the brightness diffusion value for each small patch is further detected by the detecting section, while the former is lower than the latter, the processing section encodes the brightness information for the patch by referring to a code book storing different brightness information. The encoded information is then transmitted to the decoding apparatus, together with the divided patch information. At the decoding apparatus, output image signals are reproduced from the received information by referring to the contents of a three-dimensional shape model database and a brightness information dynamic code book.
|
