Robust and Low-Rank Representation for Fast Face Identification with Occlusions

You can find the paper here.

We propose an iterative method to address the face identification problem with block occlusions. Our approach utilizes a robust representation to model contiguous errors (e.g., block occlusion) effectively based on two characteristics. The first, fits to the errors a distribution described by a tailored loss function. The second, describes the error image as structural (low- rank). We will show that this joint characterization is effective for describing errors with spatial continuity. Our approach is computationally efficient due to the utilization of the Alternating Direction Method of Multipliers (ADMM). Extensive results on representative face databases document the effectiveness of our method over existing robust representation methods with respect to both identification rates and computational time.

A special case of our fast iterative algorithm leads to the robust representation method which is normally used to handle non-contiguous errors (e.g., pixel corruption). Experimental results report significant boost on speed in this case.