稀疏正交普鲁克回归处理跨姿态人脸识别问题-计算机科学.PDFVIP

Vo1. 44 No.2 第 44 卷第2 期 计算机科学 2017 年 2 月 COMPUTER SCIENCE Feb.2017 稀疏正交普鲁克回归处理跨姿态人脸识别问题 张娟 (南京航空航天大学计算机科学与技术学院 南京 210016) 摘 要 正交普鲁克分析是一种常用的处理矩阵近似问题的技术。最近，该技术被引入到正交普鲁克回归模型中来 处理人脸姿态识别问题并取得了不错的效采。然而，这个模型对残差项使用了矩阵F 范数约束，使得模型对于一些噪 声(比如光照)非常敏感。为解决该问题，用更加鲁棒的 1 范约束替代房、始的矩阵 F 范数约束，提出稀疏正交普鲁克回 归模型。该模型可以由一个有效的交替迭代算法解决。在几个流行的人脸数据库上做了验证实验，实验结果证明该 模型可以有效地处理人脸姿态变化。 关键词 正交普鲁克分析，人脸姿态，回归模型 中固法分类号 TP312 文献标识码 A DOI 10. 11896/j. issn. 1002-137X. 2017. 02. 051 Sparse Orthogonal Procrustes Problem ßased Regression for Face Recognition with Pose Variations ZHANGjuan (College of Computer Science and Technology ,Nanjing University of Aeronautics and Astronautics ,Nanjing 210016 ,China) Abstract Orthogonal Procrustes problem (OPP) is a popular technique to deal with matrix approximation problem. Recently ,OPP was introduced into a regression model named orthogonal Procrustes problem based regression (OPPR) to handle facial pose variations and achieved interesting results. However ,OPPR performs F-norm constraint on the er- ror term ,which makes the model sensitive to the noises (i. e. , illumination variations). To address this problem ,in this paper ,the F-no口n constraint was replaced by the L -no口n constraint and the sparse orthogonal Procrustes problem 1 based regression (SOPPR) model was proposed ,which is more robust. The proposed model was then solved by an effi cient altemating iterative algorithm. Experimental results on public face databases demonstrate the effectiveness of the