| 摘要 |
FIELD: physics; image processing. ^ SUBSTANCE: proposed is a device and method of three-dimensional photorealistic simulation of a head based on an image. The method of creating a three-dimensional photorealistic model of a head involves detection of frontal and profile distinctive features in input frontal and profile images. A three-dimensional model of a head is generated through adjustment of the three-dimensional universal model. Using the detected distinctive features of the face, three-dimensional texture is generated from the input frontal and profile images and the texture is displayed on the three-dimensional model of the head. In this device and method, data obtained using a relatively cheap device, such as a digital camera, can be processed automatically, and satisfactory results can be obtained even from imperfect input data. ^ EFFECT: possibility of automatic extraction of distinctive features of a face, using a reliable algorithm for analysing a face with "quality attributable to a person". ^ 54 cl, 75 dwg |
