利用协方差矩阵法表示深度图像集的鲁棒人脸识别

Robust face recognition using block based on covariance matrix to represent depth image sets

针对深度图像集人脸识别的鲁棒性问题,提出将多幅Kinect图像作为一个图像集,Kinect捕获的原始深度数据可用于姿态估计以及人脸区域的自动裁剪。首先,将图像集划分到c个图像子集,子集中的所有图像划分为4×4的图像块;然后,将图像集中的图像模拟为图像块,按照姿势划分,每个子集使用协方差矩阵法表示;最后,在黎曼流形上模拟子集图像,为了分类,黎曼流形的每个图像子集分别学习支持向量机模型,并引入一种融合方法来合并所有图像子集的结果。在三个最大的公开Kinect人脸数据集Curtin Faces、Biwi Kinect和UWA Kinect上的实验结果验证了该方法的有效性,与其他较先进的方法相比,识别率有较大提升,标准差保持较低,对图像集数量、图像子集划分数量和空间分辨率具有较好的鲁棒性。

As the issue of robustness of face recognition based on depth image sets, this paper proposed that multiple Kinect images was being as a set of images, and used depth data captured to automatically estimate poses and crop face area. Firstly, it divided image sets into c subsets, and divided the images in all the subsets into image blocks of 4 × 4. Then, it simulated images in sets as a form of image blocks, dividing in accordance with posture, it represented each set using covariance matrix. Finally, the simulation of images in subsets is on Riemannian manifold. In order to classify, separately learnt SVM models for each image subset on the Lie group of Riemannian manifold and introduced a fusion strategy to combine results from all image subsets. It verifies the effectiveness of the proposed method on the three largest public Kinect face data sets CurtinFaces, Biwi Kinect and UWA Kinect. Compared to other advanced methods, the recognition rate has improved greater, the standard deviation is kept low, with robust to the number of image sets, image sub-setting number and spatial resolution.