Vision-Based Lip Contour Extraction and Tracking for Generating MPEG-4 Facial Animation Parameters

碩士 === 國立東華大學 === 資訊工程學系 === 94 === We propose a method for extracting lip contour using deformable templates controlled by quadratic polynomials. The proposed method combines color, shape and motion cues to estimate the contours of the outer lips. An error function is designed for merging these cue...

Full description

Bibliographic Details
Main Authors: Chih-Ming Lin, 林志明
Other Authors: Mau-Tsuen Yang
Format: Others
Language:zh-TW
Published: 2006
Online Access:http://ndltd.ncl.edu.tw/handle/x827hh
id ndltd-TW-094NDHU5392007
record_format oai_dc
spelling ndltd-TW-094NDHU53920072018-05-10T04:22:03Z http://ndltd.ncl.edu.tw/handle/x827hh Vision-Based Lip Contour Extraction and Tracking for Generating MPEG-4 Facial Animation Parameters 視覺式嘴唇輪廓偵測與追蹤—應用於產生MPEG-4臉部動畫參數 Chih-Ming Lin 林志明 碩士 國立東華大學 資訊工程學系 94 We propose a method for extracting lip contour using deformable templates controlled by quadratic polynomials. The proposed method combines color, shape and motion cues to estimate the contours of the outer lips. An error function is designed for merging these cues and a hill-climbing algorithm is applied to obtain the parameters that minimize the error function. In addition, the movements of the feature points around the lip contour are extracted based on the specification of the MPEG-4 facial animation parameters. Then the extracted parameters can be used to synthesize and animate a virtual face. Mau-Tsuen Yang 楊茂村 2006 學位論文 ; thesis 34 zh-TW
collection NDLTD
language zh-TW
format Others
sources NDLTD
description 碩士 === 國立東華大學 === 資訊工程學系 === 94 === We propose a method for extracting lip contour using deformable templates controlled by quadratic polynomials. The proposed method combines color, shape and motion cues to estimate the contours of the outer lips. An error function is designed for merging these cues and a hill-climbing algorithm is applied to obtain the parameters that minimize the error function. In addition, the movements of the feature points around the lip contour are extracted based on the specification of the MPEG-4 facial animation parameters. Then the extracted parameters can be used to synthesize and animate a virtual face.
author2 Mau-Tsuen Yang
author_facet Mau-Tsuen Yang
Chih-Ming Lin
林志明
author Chih-Ming Lin
林志明
spellingShingle Chih-Ming Lin
林志明
Vision-Based Lip Contour Extraction and Tracking for Generating MPEG-4 Facial Animation Parameters
author_sort Chih-Ming Lin
title Vision-Based Lip Contour Extraction and Tracking for Generating MPEG-4 Facial Animation Parameters
title_short Vision-Based Lip Contour Extraction and Tracking for Generating MPEG-4 Facial Animation Parameters
title_full Vision-Based Lip Contour Extraction and Tracking for Generating MPEG-4 Facial Animation Parameters
title_fullStr Vision-Based Lip Contour Extraction and Tracking for Generating MPEG-4 Facial Animation Parameters
title_full_unstemmed Vision-Based Lip Contour Extraction and Tracking for Generating MPEG-4 Facial Animation Parameters
title_sort vision-based lip contour extraction and tracking for generating mpeg-4 facial animation parameters
publishDate 2006
url http://ndltd.ncl.edu.tw/handle/x827hh
work_keys_str_mv AT chihminglin visionbasedlipcontourextractionandtrackingforgeneratingmpeg4facialanimationparameters
AT línzhìmíng visionbasedlipcontourextractionandtrackingforgeneratingmpeg4facialanimationparameters
AT chihminglin shìjuéshìzuǐchúnlúnkuòzhēncèyǔzhuīzōngyīngyòngyúchǎnshēngmpeg4liǎnbùdònghuàcānshù
AT línzhìmíng shìjuéshìzuǐchúnlúnkuòzhēncèyǔzhuīzōngyīngyòngyúchǎnshēngmpeg4liǎnbùdònghuàcānshù
_version_ 1718634908638773248