Abstract In this paper, we systematically study the effect of poorly registered faces on the training and inferring stages of traditional face recognition algorithms. We then propose a novel multiple-instance based subspace learning scheme for face recognition. In this approach, we iteratively update the subspace training instances according to diverse densities, using class-balanced supervised clustering. We test our multiple instance subspace learning algorithm with Fisherface for the application of face recognition. Experimental results show that the proposed learning algorithm can improve the robustness of current methods with poorly aligned training and testing data.
Face Mis-alignment Analysis by Multiple-Instance Subspace
Computer Vision – ACCV 2007 ; 11 ; 901-910
Lecture Notes in Computer Science ; 4844 , 11
2007-01-01
10 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Face Recognition , Recognition Rate , Training Image , Noisy Image , Baseline Algorithm Computer Science , Image Processing and Computer Vision , Computer Imaging, Vision, Pattern Recognition and Graphics , Pattern Recognition , Artificial Intelligence (incl. Robotics) , Biometrics , Algorithm Analysis and Problem Complexity
Unified Subspace Analysis for Face Recognition
British Library Conference Proceedings | 2003
|Nonparametric subspace analysis for face recognition
IEEE | 2005
|Unified subspace analysis for face recognition
IEEE | 2003
|View-subspace analysis of multi-view face patterns
IEEE | 2001
|View-Subspace Analysis of Multi-View Face Patterns
British Library Conference Proceedings | 2001
|