3D Face recognition Robust to Expression Occlusion and Poses
Ratnadeep R. Deshmukh · 2014
Face recognition has numerous applications in various identification and authentication system, but the accuracy of face recognition decreases in presence of large facial expression, occlusion and pose variations. This paper illustrates the use of scale invariant feature transform (SIFT) on 3D meshes to mode facial deformation caused by expression, occlusion and variation in poses. Here we used meshSIFT algorithm for feature extraction and sparse representation classifier for feature matching. Given a 3D face scan, its descriptors are extracted at first and then its identity can be determined by using sparse representation classifier. The proposed 3D face recognition system is robust to challenges such as large facial expressions (especially those with open mouths), large pose variations, missing parts, and partial occlusions due to glasses, hair, and so on. Our results is verified on Bosphorus Database, and I got approximate 92.46%accuracy on Bosphorus dataset.