针对动态变化背景下的人脸视频编解码问题,该文提出了一种2D-3D混合编解码系统；具体包括:(1)基于多种观测信息,在线外观模型和粒子滤波的人脸三维运动跟踪；(2)结合参数模型与肌肉模型的3D人脸动画合成；(3)基于头发检测和3D头发模型的头发合成；(4)无缝地拼接前景的三维编解码结果和背景的二维编解码结果.在极低码率下,客观实验表明,该系统在编码效率和解码质量上有较好的综合优势.主观实验表明,该系统的解码结果在脸部具有较高的辨识度.%For facial video coding/decoding with dynamic background, a 2D/3D mixed system was proposed, it includes: (1)multi-measurements and online appearance models were applied to track 3D facial motion from video in framework of particle filter; (2)3D facial animation was produced combining parameterized model and muscular model; (3)3D hair was synthesized based on hair detection and 3D hair model; (4)3D coding/decoding result of foreground and 2D coding/decoding result of background were stitched seamlessly.At ultra-low bit-rate, the object experiment confirmed the comprehensive advantage between coding efficiency and decoding quality of this system, the between subjects experiment indicated the suitability of subjective face identification by it.
展开▼
