正则化最近点优化图像集匹配的三维人脸识别

针对传统的三维人脸识别算法受光照、表情、姿态及遮掩等变化而影响识别性能的问题，提出了一种基于正则化最近点优化图像集匹配算法。将图库图像集和探针图像集建模成正则化仿射包，利用迭代器自动确定两个图像集间的正则化最近点；利用最近子空间分类器最小化正则化最近点；根据正则化最近点之间的欧氏距离及结构计算RNP集之间的距离，利用最近邻分类器完成人脸的识别。在Honda/UCSD、BU4DFE两大视频人脸数据库上的实验验证了该算法的有效性及可靠性，实验结果表明，相比其他几种较为先进的三维人脸识别算法，该算法取得了更好的识别效果，同时，大大减少了训练及测试总完成时间。

Image set matching optimized by RNP for three-dimensional face recognition

The recognition performance of traditional three-dimensional face recognition algorithms is impacted by variation of illustration, expression, pose and occlusion seriously, for which image sets matching algorithm optimized by regularized nearest points is proposed. Gallery image sets and probe image sets are modeled as regularized affine hulls, and iterator is used to conform regularized nearest points between the two sets. Recent subspace classifier is used to minimize the regu-larized nearest points. Distance between RNP sets is calculated by Euclidean distance and structure of RNP and nearest neighbor classifier is used to finish face recognition. The effectiveness and reliability of proposed algorithm have been ver-ified by experiments on the three video face databases Honda/UCSD, BU4DFE. Experimental results show that proposed algorithm has better recognition efficiency, less training time and testing time than several advanced three-dimensional face recognition algorithms.