基于特征分块的快速三维人脸重建

Fast 3D Face Reconstruction Based on Feature Division

针对传统三维人脸重建算法精度低且难以满足实际应用的缺陷,提出一种基于特征分块的快速三维人脸重建算法。该算法首先利用基于平面模板的重采样法实现三维数据规范化;其次,根据人脸特征对人脸分块,根据主动形状模型算法进行面部特征定位。然后,利用基于形变模型算法与径向基函数插值相结合,实现人脸三维重建,并将该算法应用到人脸识别中。实现表明,此算法可快速实现特定正面二维人脸的三维重建,并具有较高重建精度和识别率。

The traditional algorithm of 3D face reconstruction is inefficient and it is difficult to meet the requirements of practical application. To address this problem,a fast feature-division-based 3D face reconstruction algorithm is proposed. First,a 2D template-based alignment algorithm is developed to process the correspondence between faces automatically. Second,Each face is divided based on feature division,and an improved ASM algorithm is proposed to locate the feature points in the 3D and 2D face images. Then,every facial feature slices shape is reconstructed by a PCA-based sparse morphable mode and and then uses radial basis function to remedy the 3D data. Finally,the algorithm is applied to 3D face recognition. Experimental results show that the presented algorithm obtains faster speed to reconstruction 3D face from the 2D frontal face,keeping a quite high accuracy results and having a high recognition rate.