BROWSE
Detail
Conference Paper
Hand Segmentation for Optical See-through HMD Based on Adaptive Skin Color Model Using 2D/3D Images
Cited - time in 
Share 
Share
- Authors
- Gisu Heo, Dong-Woo Lee, Sungyong Shin, Gague Kim, Hyeongcheol Shin
- Issue Date
- 2014-08
- Citation
- International Conference on Machine Vision and Machine Learning (MVML) 2014, pp.1-5
- Language
- English
- Type
- Conference Paper
- Abstract
- In this paper, we propose a robust hand segmentation method based on adaptive skin color histogram model using the fusion of 2D/3D images for bare hand interaction in the optical see-through head-mounted-display (OHMD). Initially, the hand area detection is performed with depth information from stereo vision. The adaptive skin color histogram models are created with a chromaticity-based constraint to select pixels in a scene for updating a dynamic skin color model under changing illumination or cluttered background conditions. Experiment results show that the proposed method can accurately segmented the hand under various environmental conditions in the OHMD.