摘要 |
PROBLEM TO BE SOLVED: To provide a means for deciding which image is to be used to generate each portion for texture rendering because all the surfaces of an object can not be seen from any viewpoint and whichever a three-dimensional object is. SOLUTION: An image processing apparatus (2) is disclosed in which input images are processed to generate texture map data for texture rendering a generated three-dimensional computer model of object(s) appearing in the images. In order to select portions of the images utilized, confidence data is generated indicative of the extent of each portion of the surface of a model is visible in each of the images. The images are combined utilizing this confidence data, where image data representative of different spatial frequencies are then blended in different ways utilizing the confidence data.
|