Title :
Multi-view face pose estimation based on supervised ISA learning
Author :
Li, Stan Z. ; Peng, Xian Huan ; Hou, Xin Wen ; Zhang, Hong Jiang ; Cheng, Qian Sheng
Author_Institution :
Beijing Sigma Center, Microsoft Res. Asia, Beijing, China
Abstract :
Independent subspace analysis (ISA) is able to learn view-subspaces unsupervisedly from (view-unlabeled) multi-view face examples (S.Z. Li et al., 2001). We explain underlying reasons for the emergent formation of ISA view-subspaces. Based on the analysis, we present a supervised method for more effective learning of view-subspace, assuming that view-labeled face examples are available. The models thus learned give more accurate pose estimation than those obtained with the unsupervised ISA
Keywords :
face recognition; learning (artificial intelligence); optimisation; ISA view-subspaces; face recognition system; independent subspace analysis; multi-view face examples; multi-view face pose estimation; supervised ISA learning; supervised method; Character generation; Face recognition; Instruction sets;
Conference_Titel :
Automatic Face and Gesture Recognition, 2002. Proceedings. Fifth IEEE International Conference on
Conference_Location :
Washington, DC
Print_ISBN :
0-7695-1602-5
DOI :
10.1109/AFGR.2002.1004140
