Face Detection and Tracking in Video Sequence using Fuzzy Geometric Face Model and Mean Shift
Prakash S. Hiremath, Manjunath Hiremath, R. Mahesh · 2013
Abstract: Humans make use of face as an important cue for identifying people. This makes automatic face detection very crucial from the point of view of a wide range of commercial and law enforcement applications. While traditional face detection is typically based on still images, face detection and tracking from video sequences has become prominent research domain. In this paper, we propose a novel algorithm which segments the face region in video images using fuzzy geometric face model in key frame. The mean shift is used to track the face along the video sequence. Contrary to current techniques that are based on huge learning databases and complex algorithms to get generic face models, the proposed method handles simple face detection and tracking approach. The proposed method is implemented and evaluated with numerous experiments on videos containing large variations of head motion, light condition, and expressions. The experimental results show that the proposed method is effective in detecting and tracking faces in videos. Key words: Face detection, fuzzy geometric face model, mean shift, face tracking.