3D-2D mapping based large-pose face recognition

Abstract:

To solve the problem of low accuracy of large pose face recognition in real scenes， a large pose face recognition algorithm framework is proposed based on the 3D-2D mapping focusing on the exploration of the key technologies of large pose face recognition from the perspective of data augmentation. Different from the current face recognition algorithm based on 3D point cloud data， the proposed method mainly uses 3D face data to enrich the posture information via the data expansion method by 3D-2D mapping. The specific posture face feature extraction models are trained and integrated into the unified large pose face recognition framework. The key of the proposed method is to use registered 3D face information to assist 2D face multi-pose recognition. It is easy to integrate existing 2D face recognition methods into the proposed framework. Experiments show that the method proposed in this paper can effectively improve the accuracy of large pose face recognition in unconstrained real scenes without increasing the significant computational load and provides a new idea for the current 3D information to solve the 2D face recognition problem.