Comparing and combining depth and texture cues for face recognition

Depth and texture play complementary roles in the coding of faces. While most existing face recognition methods are based on texture cues from intensity images due to their ease of acquisition, more accurate face recognition can be achieved by exploiting both modalities. To evaluate this claim, we have designed a pattern classifier for three different inputs: (1) depth map, (2) texture map, and (3) both depth and texture maps of the facial surface. Two existing face recognition methods are considered for this task, FaceIt technology and FisherFaces. Recognition performance is evaluated based on face data of 185 subjects, captured with the structured-light-based Rainbow250 3D camera.