Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/40702
Title: Realistic and expressive computer facial animation
Authors: Chan, Choong Seng.
Keywords: DRNTU::Engineering
Issue Date: 2010
Abstract: Computer facial animation still remains a very challenging topic within the computer graphics community. In this project, a realistic and expressive computer facial animation system is developed by automated learning from Vicon system facial motion capture data. Data captured from real speaker are used which would make the mapped facial animation more natural and lifelike. The dimensions of facial motion capture data with different emotions are collected using Vicon Nexus are reduced using dimensionality reduction techniques. The purpose of reduce the dimensions of the data is to prepare them to be used in the construction of the Phoneme-Independent Expression Eigenspace (PIEES) in the modeling stage. Programs are written using Matlab to perform dimensionality reduction techniques on the data. The quality of the resulting data is evaluated by computing the Root Mean Square Error. After that, the resulting data from Matlab is converted to .C3D format using C3Deditor. Ultimately, the emotions data in .C3D format are mapped to a 3D animated face using Autodesk MotionBuilder 2009. Our experiments demonstrate that the approach presented can effectively synthesize realistic expressive facial animation. The approach presented can be used for various applications to generate natural expressive and realistic facial animation, for example, generating a personal avatar for different users.
URI: http://hdl.handle.net/10356/40702
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
Chan Choon Seng_FYP.pdf
  Restricted Access
8.34 MBAdobe PDFView/Open

Page view(s)

212
checked on Sep 28, 2020

Download(s)

5
checked on Sep 28, 2020

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.