Abstract: This paper analyzes the feature of 3D face recognition, and presents a 3D face recognition algorithm based on 2D Gabor wavelet feature. This algorithm uses 2D Gabor wavelet feature to describe features on human face accurately and stably. The 3D face recognition is performed by three steps: 3D face model reconstruction, pattern matching of reconstructed 3D face model and Linear Discriminates Analysis of matched 3D face model. The experimental result based on ORL and UMIST databases shows that this method has excellent performances.

Key words: 2D Gabor wavelet feature, 3D face recognition, face appearance model, model matching

摘要： 分析三维人脸识别技术，提出一种基于Gabor小波特征的三维人脸识别算法。该算法采用二维Gabor小波特征精确且稳定地描述人脸特征，重建三维人脸模型并对其进行模板匹配，对匹配后的三维人脸模型进行线性判别分析。对基于ORL和UMIST两个人脸数据库的实验结果表明，该算法性能优良。

关键词: 二维Gabor小波特征, 三维人脸识别, 人脸表面模型, 模型匹配

CLC Number: 

• TP391.4