Analysis and synthesis of facial image sequences in model-based image coding

摘要: This paper proposes new methods for analyzing image sequences and updating textures of the three-dimensional (3-D) facial model. It also describes a method synthesizing various expressions. These three are key technologies model-based coding system. The input analysis technique directly robustly estimates 3-D head motions expressions without any two-dimensional (2-D) entity correspondences. resolves 2-D correspondence mismatch errors provides quality reproduction original images by fully incorporating synthesis rules. To verify algorithm, performs quantitative subjective evaluations. presents two texture model to improve synthesized images. first focuses on parts with large change brightness according reducing transmission bit rates. second all changes caused as well rates estimated update methods. For output images, it rules that simulate muscular actions because muscles cause more easily synthesize high-quality represent >