Please use this identifier to cite or link to this item: http://scholarbank.nus.edu.sg/handle/10635/29955
Title: Facial Expression Recognition: Fusion of A Human Vision System Model and A statistical framework
Authors: GU WENFEI
Keywords: Facial Expression Recognition, Human Vision System, Local Gabor Features, Radial Encoding, Classifier Synthesis, Composite Orthonormal Basis
Issue Date: 18-May-2011
Source: GU WENFEI (2011-05-18). Facial Expression Recognition: Fusion of A Human Vision System Model and A statistical framework. ScholarBank@NUS Repository.
Abstract: Automatic facial expression recognition from still face (color and gray-level) images is acknowledged to be complex in view of significant variations in the physiognomy of faces with respect to head pose, environment illumination and person-identity. Even assuming illumination and pose invariance in face images, recognition of facial expressions from novel persons always remains an interesting and also challenging problem. With the goal of achieving significantly improved performance in expression recognition, the proposed new algorithms, combining bio-inspired approaches and statistical approaches, involve (a) the extraction of contour-based features and their radial encoding; (b) a modification of HMAX model using local methods; and (c) a fusion of local methods with an efficient encoding of Gabor filter outputs and a combination of classifiers based on PCA and FLD. In addition, the sensitivity of existing expression recognition algorithms to facial identity and its variations is overcome by a novel composite orthonormal basis that separates expression from identity information. Finally, by way of bringing theory closer to practice, the proposed facial expression recognition algorithm has been efficiently implemented for a web application.
URI: http://scholarbank.nus.edu.sg/handle/10635/29955
Appears in Collections:Ph.D Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
GuWF.pdf4.52 MBAdobe PDF

OPEN

NoneView/Download

Page view(s)

352
checked on Dec 11, 2017

Download(s)

334
checked on Dec 11, 2017

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.