Publication Details
Face Representation and Tracking Using Gabor Wavelet Networks
Image, Object representation, Feature-based, Template-based, 2D Gabor wavelet functions, Gabor Wavelet Networks, Facial features, Tracking, Facial expressions, Parameterized models of optical flow, Principal Component Analysis
This work presents a one of the approaches to human gesture recognition. A method where a discrete face template is represented by linear combination of the 2D odd-Gabor wavelet functions (Gabor Wavelet Network) is proposed. Using wavelet networks, an effective face representation and tracking method is achieved that is robust to changes and deformations of the face image. Possibilities of cooperation between a GWN and others methods for robust facial expressions recognition are also discussed, for example: Principal Component Analysis (PCA) and parameterized models of optical flow.
@INPROCEEDINGS{FITPUB7366, author = "Michal \v{S}pan\v{e}l and Pavel Zem\v{c}\'{i}k", title = "Face Representation and Tracking Using Gabor Wavelet Networks", pages = "163--169", booktitle = "Proceedings of the Central European Seminar on Computer Graphics", year = 2003, location = "Budmerice, SK", language = "english", url = "https://www.fit.vut.cz/research/publication/7366" }