PROJECT TITLE :

Robust Face Recognition from Multi View Videos - 2014

ABSTRACT:

Multiview face recognition has become an active research area in the last few years. In this paper, we have a tendency to gift an approach for video-primarily based face recognition in camera networks. Our goal is to handle pose variations by exploiting the redundancy within the multiview video data. However, in contrast to traditional approaches that explicitly estimate the pose of the face, we propose a completely unique feature for sturdy face recognition in the presence of diffuse lighting and create variations. The proposed feature is developed using the spherical harmonic illustration of the face texture-mapped onto a sphere; the feel map itself is generated by back-projecting the multiview video data. Video plays an important role in this state of affairs. 1st, it provides an automatic and economical method for feature extraction. Second, the information redundancy renders the recognition algorithm a lot of robust. We have a tendency to live the similarity between feature sets from totally different videos using the reproducing kernel Hilbert space. We have a tendency to demonstrate that the proposed approach outperforms traditional algorithms on a multiview video database.