Very low bit rate face video compression using linear combination of 2D face views and principal components analysis

Citation
I. Koufakis et Bf. Buxton, Very low bit rate face video compression using linear combination of 2D face views and principal components analysis, IMAGE VIS C, 17(14), 1999, pp. 1031-1051
Citations number
49
Categorie Soggetti
AI Robotics and Automatic Control
Journal title
IMAGE AND VISION COMPUTING
ISSN journal
02628856 → ACNP
Volume
17
Issue
14
Year of publication
1999
Pages
1031 - 1051
Database
ISI
SICI code
0262-8856(19991117)17:14<1031:VLBRFV>2.0.ZU;2-G
Abstract
We present an image synthesis technique for very low bit rate encoding of f aces for videoconferencing applications over the Internet or mobile communi cation networks. The proposed scheme represents each new face view as a lin ear combination of three basis face views of the same person. The backgroun d of the target face view is also encoded by the linear combination method using the basis views used for the face area. However, changes in the eyes and mouth regions cannot be encoded by the Linear combination scheme. These regions are encoded using principal components analysis with three trainin g sets of eyes and mouth images of the same person. The scheme results in a very compact representation of the face video data. Good quality images ar e reconstructed with an estimated bit rate varying from 1384 to 1620 bits/f rame. (C) 1999 Elsevier Science B.V. All rights reserved.