A Framework for Face Recognition from Video Sequences Using GWN and Eigenfeature Selection
TE deCampos, FS Feris, RM Cesar-Jr · 2000
. This work describes an ongoing project related to a new method to perform face recognition from video sequences. Faces are detected and tracked in a video sequence using Gabor wavelet networks. This process also allows locating and extracting facial feature regions around the eyes, nose and mouth. A modified Karhunen-Love transform, defined with the aid of an automatic feature selection procedure, is used for feature extraction and face recognition. Several preliminary results are discussed. Keywords: face detection, tracking of facial features, face recognition, Gabor Wavelet Networks, Principal Components Analysis and Feature Selection. 1