Large Scale Learning of Active Shape Models
Image Processing, 2007. ICIP 2007. IEEE International Conference(2007)
摘要
We propose a framework to learn statistical shape models for faces as piecewise linear models. Specifically, our methodol- ogy builds upon primitive active shape models(ASM) to han- dle large scale variation in shapes and appearances of faces. Non-linearities in shape manifold arising due to large head rotation cannot be accurately modeled using ASM. Moreover overly general image descriptor causes the cost function to have multiple local minima which in turn degrades the qual- ity of shape registration. We propose to use multiple overlap- ping subspaces with more discriminative local image descrip- tors to capture larger variance occurring in the data set. We also apply techniques to learn distance metric for enhancing similarity of descriptors belonging to the same class of shape subspace. Our generic algorithm can be applied to large scale shape analysis and registration.
更多查看译文
关键词
image registration,piecewise linear techniques,active shape model,cost function,distance metric learning,face alignment problem,facial feature localization,generic algorithm,image descriptor,large scale shape analysis,multiple overlapping subspaces,nonlinear shape manifold,piecewise linear model,shape registration,statistical shape model learning,Active Shape Models,Anderson Darling Statistics,Relevance Component Analysis,SIFT
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络