Lip contour extraction using level set curve evolution with shape constraint

Authors:
Jae Sik Chang;Eun Yi Kim;Se Hyun Park
Affiliations:
Computer Science Dept., University of California, Santa Barbara, CA;School of Internet and Multimedia, Next-Generation Innovative Technology Research Institute, Konkuk Univ., Seoul, Korea;School of Computer and Communication, Daegu Univ., Daegu, Korea
Venue:
HCI'07 Proceedings of the 12th international conference on Human-computer interaction: intelligent multimodal interaction environments
Year:
2007

Citing 6
Cited 0

Fronts propagating with curvature-dependent speed: algorithms based on Hamilton-Jacobi formulations

Journal of Computational Physics
Region Competition: Unifying Snakes, Region Growing, and Bayes/MDL for Multiband Image Segmentation

IEEE Transactions on Pattern Analysis and Machine Intelligence
Region Tracking via Level Set PDEs without Motion Computation

IEEE Transactions on Pattern Analysis and Machine Intelligence
Active contours without edges

IEEE Transactions on Image Processing
Lip image segmentation using fuzzy clustering incorporating an elliptic shape function

IEEE Transactions on Image Processing
Accurate and quasi-automatic lip tracking

IEEE Transactions on Circuits and Systems for Video Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this work, a novel method for lip contour extraction based on level set curve evolution is presented. This method takes not only color information but also lip contour shape constraint represented by a distance function between the evolving curve and parametric shape model. In this method, the curve is evolved by minimizing an energy function that incorporates shape constraint function as internal energy, while previous curve evolution methods use a simple smoothing function. The new shape constraint function prevents the curve from evolving to arbitrary shapes occurred due to weak color contrast between lip and skin regions. Comparisons with other method are conducted to evaluate the proposed method. It showed that the proposed method provides more accurate results than other methods.