29 December 2000 Object segmentation using skin color and motion information
Author Affiliations +
Abstract
This paper presents an effective object segmentation method for object-oriented coding. The process is composed of facial region detection using skin color and changed region detection using motion information. The image is then segmented between moving objects and background using motion estimation. The facial regions are detected using skin color, the data for which is obtained from the u, v image. This data is then used as a threshold to segment the facial region. The image is also segmented into changed regions using motion information. After combining these results, the final image is segmented between moving objects and background. This method is more efficient due to two factors, color and motion. Experimental results show that the proposed method can significantly improve the image quality.
© (2000) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Eui-Yoon Chung, Eui-Yoon Chung, Ho-Keun Lee, Ho-Keun Lee, Hyung Suk Kim, Hyung Suk Kim, Yeong-Ho Ha, Yeong-Ho Ha, } "Object segmentation using skin color and motion information", Proc. SPIE 4310, Visual Communications and Image Processing 2001, (29 December 2000); doi: 10.1117/12.411854; https://doi.org/10.1117/12.411854
PROCEEDINGS
9 PAGES


SHARE
Back to Top