Portrait drawing from corresponding range and intensity images

摘要: We propose a real-time rendering system for automatically creating caricature drawing, i.e., an exaggerated portrait, of human face, based on simultaneous use range image (or 3D mesh) and registered photograph the same face. Combining these information sources provides complementary information. Significant geometric lines such as occluding contours suggestive are extracted from data, while textured areas corresponding to shading features photograph. These combined, then distorted produce final caricature. The output may be produced using choice non-photorealistic styles. Our method works well low resolution images; it is fast enough allow viewpoint chosen in real time. combines significant lines, areas, optional shading, giving pleasing result which preserves not only shape cues description, but also other essential visual characteristics facial that cannot deduced geometry alone.