一种基于稠密SIFT特征对齐的稀疏表达人脸识别算法

Improved Sparse Representation Algorithm for Face Recognition Via Dense SIFT Feature Alignment

该文针对人脸图像受到非刚性变化的影响,如旋转、姿态以及表情变化等,提出一种基于稠密尺度不变特征转换(SIFT)特征对齐(Dense SIFT Feature Alignment,DSFA)的稀疏表达人脸识别算法。整个算法包含两个步骤:首先利用DSFA方法对齐训练和测试样本;然后设计一种改进的稀疏表达模型进行人脸识别。为加快DSFA步骤的执行速度,还设计了一种由粗到精的层次化对齐机制。实验结果表明:在ORL,AR和LFW 3个典型数据集上,该文方法都获得了最高的识别精度。该文方法比传统稀疏表达方法在识别精度上平均提高了4.3%,同时提高了大约6倍的识别效率。

In order to address the non-rigid deformation （e.g., misalignment, poses, and expression） of facial images, this paper proposes a novel sparse representation face recognition algorithm using Dense Scale Invariant Feature Transform （SIFT） Feature Alignment （DSFA）. The whole method consists of two steps： first, DSFA is employed as a generic transformation to roughly align training and testing samples; and then, input facial images are identified based on proposed sparse representation model. A novel coarse-to-fine scheme is designed to accelerate facial image alignment. The experimental results demonstrate the superiority of the proposed method over other methods on ORL, AR, and LFW datasets. The proposed approach improves 4.3% in terms of recognition accuracy and runs nearly 6 times faster than previous sparse approximation methods on three datasets.