Sciweavers

BMVC
2010

Skin Color Weighted Disparity Competition for Hand Segmentation from Stereo Camera

13 years 9 months ago
Skin Color Weighted Disparity Competition for Hand Segmentation from Stereo Camera
Hand segmentation is the basis of many vision-based hand gesture applications in human computer interaction (HCI). This paper proposes a novel method of skin color weighted disparity competition to incorporate the skin color and depth cues for hand segmentation in clutter environment from stereo camera. The basic idea is that the hand disparity will be apparently different from most of the background disparities. Note that we refer to the scene except the user as the background. The proposed method validates a hand pixel by verifying whether the pixel satisfies the disparity limit of the hand more than of the background. Its novelty lies in incorporating the skin color probability to weight the disparity competition to improve the robustness of validation and the accuracy of boundary location. Experiments demonstrate the efficiency of the proposed method.
Qi Wang, Xilin Chen, Wen Gao
Added 10 Feb 2011
Updated 10 Feb 2011
Type Journal
Year 2010
Where BMVC
Authors Qi Wang, Xilin Chen, Wen Gao
Comments (0)