Robust real time face recognition and tracking on gpu using fusion of rgb and depth image

摘要

This paper presents a real-time face recognition system using kinect sensor.The algorithm is implemented on GPU using opencl and significant speedimprovements are observed. We use kinect depth image to increase the robustnessand reduce computational cost of conventional LBP based face recognition. Themain objective of this paper was to perform robust, high speed fusion basedface recognition and tracking. The algorithm is mainly composed of three steps.First step is to detect all faces in the video using viola jones algorithm. Thesecond step is online database generation using a tracking window on the face.A modified LBP feature vector is calculated using fusion information from depthand greyscale image on GPU. This feature vector is used to train a svmclassifier. Third step involves recognition of multiple faces based on ourmodified feature vector.