A Key Point Extraction Method of Lip Contours Based on Jumping Snake

DOI

10.2991/icmse-15.2015.72How to use a DOI?

Keywords

lip segmentation; key point extraction; color space; jumping snake

Abstract

Lip segmentation and key point extraction from human mouth image are essential steps in lip-reading applications. In this paper, a method is proposed to improve key point extraction. First, mouth is detected according to the structure and proportion of the face. Second, two corners of the mouth are detected and extracted. Third, we study many color spaces and find more appropriated color space and gradients for outer and inner lip contours. And at the last, we use the jumping snake algorithm to extract key points along outer and inner lip contours. The proposed method gives promising results but still has little inaccuracy with inner contours.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Yuanyao Lu
AU  - Song Geng
PY  - 2015/12
DA  - 2015/12
TI  - A Key Point Extraction Method of Lip Contours Based on Jumping Snake
BT  - Proceedings of the 2015 6th International Conference on Manufacturing Science and Engineering
PB  - Atlantis Press
SP  - 394
EP  - 400
SN  - 2352-5401
UR  - https://doi.org/10.2991/icmse-15.2015.72
DO  - 10.2991/icmse-15.2015.72
ID  - Lu2015/12
ER  -