3D Face Recognition Algorithm Based on Discriminant Analysis Using Composite Features
-
摘要: 针对人脸识别中特征表示与提取问题, 提出了一种新颖的基于组合特征判别分析的三维人脸识别算法.该算法首先使用基于非均匀网格重采样的方法对所有三维人脸做规格化处理, 使三维人脸具有统一的点数和拓扑结构;其次, 以先分段、再重叠的形式将原本一维向量表示组织为二维矩阵表示, 然后使用二维线性判别分析方法 (2DLDA) 对获得的数据进行特征抽取.这种方法在避免图像信息的丢失、增加组合特征的同时, 理论上也能避免单纯使用线性判别分析 (LDA) 进行特征抽取时容易出现的小样本问题.在BJUT-3D大规模三维人脸数据库上的实验表明, 本方法取得了良好的识别效果.Abstract: This paper proposes an approach of face recognition using composite features.This approach first aligned 3D faces based on non-uniform mesh re-sampling which result the uniform number of vertex and topology structure and kept in a vector.Second, we changed the vector to the matrix by division and overlap, and then 2DLDA suggests a feature selection strategy to select the most discriminative features from the corner.This approach not only avoided face information missing, increased the number of composite feature, but also avoided the small sample size problem in theory.Experimental results for 3D face data set from BJUT-3D face database have demonstrated the performance of our algorithm.
-
Keywords:
- face recognition /
- feature extraction /
- discriminant analysis
-
-
[1] ZHAO W, CHELLAPPA R, ROSENFELD A.Face recognition:a literature survey[C]∥Proceeding of ACM ComputingSurveys.New York:ACM Press, 2003:399-458.
[2] MOSES Y, ADINI Y, ULLMAN S.Face recognition:the problem of compensating for illumination changes[J].IEEETransactions on PAMI, 1997, 19 (7) :721-732.
[3] BOWYER K W, CHANG K, FLYNN P.A survey of approaches and challenges in 3D and multi-modal 2D+3D facerecognition[J].Computer Vision and Image Understanding, 2006, 101 (1) :1-15.
[4] BELHUMEUR P N, HESPANHA J P, KRIEGMAN D J.Eigenfaces vs.fisherfaces:recognition using class specific linearprojection[J].IEEE Transactions on PAMI, 1997, 19 (7) :711-720.
[5] SUN Yan-feng, WANG Jun, YIN Bao-cai.A3D face recognition algorithm based on nonuniform re-sampling correspondence[C]∥Advances in Visual Computing-Third International Symposium, ISVC 2007, proceedings.Lake Tahoe:SpringerVerlag, 2007, 407-416.
[6] CHEN L, LIAO H, KO M, et al.A new LDA-based face recognition system which can solve the small simple size problem[J].Pattern Recognition, 2000, 33 (10) :1713-1726.
[7] CEVIKAPL H, NEAMTUM, WILKE S, et al.Discriminative common vectors for face recognition[J].IEEE Transactions onPAMI, 2005, 27 (1) :4-13.
[8] XIONG H, SWAMY M N S, AHMAD M O.Two-dimensional FLD for face recognition[J].Pattern Recognition, 2005, 38 (7) :121-1124.
[9] JING X Y, WONG X S, ZHANG D, Face recognition based on 2D fisherface approach[J].Pattern Recognition, 2006, 39 (4) :707-710.
[10] NOUSHATHS, KUMAR G H, SHIVAKUMARA P. (2D) 2LDA:An efficient approach for face recognition[J].PatternRecognition, 2006, 39:1396-1400.
[11] 尹宝才, 何晏晏, 孙艳丰, 等.三维人脸的非均匀重采样对齐技术[J].北京工业大学学报, 2007, 33 (2) :213-218.YIN Bao-cai, HE Yan-yan, SUN Yan-feng, et al.Nonuniform resampling based on method for pixel-wise correspondencebetween 3D faces[J].Journal of Beijing University of Technology, 2007, 33 (2) :213-218. (in Chinese) [12] KIM C, CHOI C H.Image covariance-based subspace method for face recognition[J].Pattern Recognition, 2007, 40 (5) :1592-1604.
计量
- 文章访问数: 12
- HTML全文浏览量: 0
- PDF下载量: 7
下载:
