Face Inpainting with Local Linear Representations

A numberof investigatorshave had success usingdomainspecific priorknowledgeto pro- duce improved superresolution images of faces ("hallucinating faces"). These efforts ad- dress the scenario where a face image is obtained from a low-resolutioncamera. A related but less studied problem occurs when the missing information is the result of occlusion rather than low camera resolution, as in the case when a person is wearing sunglasses. Recently Hwang and Lee (14) introduced the first algorithm for solving this reconstruc- tion "inpainting" problem. In the current work we report results of a psychological study that provides independent evidence regarding the validity of the face reconstruction task, and we demonstrate an improved reconstruction approach using a positive, local linear representation. The positive, local mixture operates on real-world images without manual intervention in many cases, and provides demonstrably lower reconstruction error than is obtainable with a global representation.