Robust Face Tracking using Improved Mean Shift Algorithm

Abstract

Face tracking is important system for robotic. The problem is that it is less resistant to occlusion and noise. Robust face tracking algorithm in this paper using improved mean shift algorithm is proposed. This algorithm consist of face detection and tracking. The face identification is done by the Viola-Jones algorithm. Then, Mean Shift algorithm is done to track face as a target and improve robustness from occlusion. CBWH is added in face as target to reduce noise. The experimental results prove that it is robust from noise background and occlusion..

Self scale estimation of the tracking window merged with adaptive particle filter tracker

“…Deterministic approaches follow targets cross frames by searching iteratively for area most similar to the target window area via maximizing measure between those areas. A typical deterministic method is mean shift [5], it proposes a face tracking algorithm with an improved implementation of mean shift and identification of the face was based on Viola and Jones algorithm, they have also used corrected background weighted histogram to reduce noise on face on the process. More deterministic approaches were adopted by integrating other feature description such as texture feature [6], scale-invariant feature transform (SIFT) [7], cross-bin metric [8], tracking by detection based on Hungarian algorithm [9], discriminative correlation filters [10]- [12], online multiple instance learning [13] and positive and negative learning bootstrapping binary classifiers [14].…”