Continuously adaptive mean-shift (CAMShift) is an efficient and light-weight tracking algorithm firstly developed based on mean-shift to track human face in a perceptual user interface. CAMShift tracks a target by searching for the most similar areas to the target region in frames of a video sequence in regard to its reference color histogram. While color based CAMShift is suitable for tracking targets in simple cases, it fails to track objects in more complex situations. In this paper we review our low cost extension to improve the traditional CAMShift algorithm. Combining the original algorithm with simple motion segmentation techniques, we proposed an improved CAMShift algorithm to cope with CAMShift`s tracking problems. We have evaluated the efficiency of our approach through analyzing our tracking results. We have compared our results with other tracking algorithms in various tracking scenarios.
Rights and permissions | |
![]() | This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |