Paper Reading - 先进人机通信技术联合实验室.pptVIP

;Papers;;Outline;Outline;Author Information;Outline;Abstract;Outline;提出了一种从样本图像自动生成NPR人脸动画的新方法 首先给定各种人脸表情的手绘图集 6 different mouths 4 pairs of eyes 1 overall head 对某个指定用户一次训练后， 得出目标人脸和手绘元素相似表情之间的尺长 算法跟踪发送方的表情，提取为几个参数发送给渲染器 渲染器使用多种数据插值和扭曲方法，从合适的画集中合成出动画形象;系统需求 用可变形控制线标注好的手绘图像集 划分为可以被扭曲和混合的三部分 Mouth Eyes Background head images 要求有着可辨别的表情和唇部位置 6个口型和4对眼睛的图像即足以 在远程会议中产生好的结果;What does this paper do?(3/4);What does this paper do?(4/4);Outline;Motivation;Outline;How to do?;How to do?;Tracking(1/3);;Tracking(3/3);How to do?;Rendering;Rendering;Expression mapping for mouth(1/2);Expression mapping for mouth(2/2);Mouth interpolations;Mouth warping and blending;Rendering;Warping(1/2);Warping(2/2);Outline;Implemented hardware 450MHz Pentium III processor A high-end PC graphics card The tracker produces 10 8-bit integers per frame, current bandwidth requirements are 2400 baud for 30 frames per second. At 10 fps, the bandwidth requirement drops to a miserly 800 baud. ;Experiment(2/3);Experiment(3/3);Outline;Appendix;;Outline;Outline;Ankur Patel Experience 2003-2005, M.S. , Electrical and Computer Engineering at Rutgers, The State University of New Jersey-New Brunswick. Jan.2005-Jul.2008, Research and Development Engineer at CyberExtrude. Aug.2008-present , Ph.D. student in the Department of Computer Science at The University of York. Research Computer Vision(3DMM, SFS) Publication A. Patel and W.A.P. Smith. “Shape-from-shading driven 3D Morphable Models for Illumination Insensitive Face Recognition.” In Proc BMVC. 2009. ;Author Information(2/2);Outline;Abstract;Outline;平均网格;Outline;How to do?;How to do?;A thin-plate spline warp;Use Procrustes analysis to remove pose effects without having to choose a reference face. Shape Alignment: Mean shape estimate: Fix the length of the mean shape at each iteration: Using a 3D similarity transform to align the sample to the current estimate of the mean shape: where map a sample onto the mean shape vecto