Lip contour tracking using an improved ASM considering edge, color and shape features

碩士 === 國立東華大學 === 資訊工程學系 === 97 === Lip tracking is a popular research topic in the field of computer vision. In this thesis, lip contours are located and tracked based on Active Shape Models (ASM). In addition, we combine and exploit the edge, color and shape features in an input face video to impr...

Full description

Bibliographic Details
Main Authors: Jheng-Ting Huang, 黃琤婷
Other Authors: Mau-Tsuen Yang
Format: Others
Language:zh-TW
Published: 2008
Online Access:http://ndltd.ncl.edu.tw/handle/04484954847269980864
Description
Summary:碩士 === 國立東華大學 === 資訊工程學系 === 97 === Lip tracking is a popular research topic in the field of computer vision. In this thesis, lip contours are located and tracked based on Active Shape Models (ASM). In addition, we combine and exploit the edge, color and shape features in an input face video to improve the traditional ASM in that the detection relies only on pre-processed gray level images. The experimental results show that the proposed method tracks inner and outer lip contours more accurate compared to the traditional ASM.