Abstract
In face recognition, image resolution is an important factor which has a great influence on the recognition rate. In traditional face recognition for low-resolution images, face interpolation/super-resolution is usually performed first, and the constructed high-resolution face image will then pass through a face recognition system, which includes feature extraction and classification. To achieve a more efficient and accurate approach, we propose a new method of Gabor-Feature Hallucination, which predicts the high-resolution Gabor features from the low-resolution Gabor features directly by using linear regression and Generalized Canonical Correlation Analysis. Then, the low-resolution features in the projected Generalized Canonical Correlation space and the predicted high-resolution Gabor features are adopted for face classification. Our algorithm can therefore avoid performing interpolation/super-resolution and high-resolution Gabor feature extraction. Experimental results show that the proposed method has a superior recognition rate and efficiency to the traditional methods.
Original language | English |
---|---|
Title of host publication | 2011 International Symposium on Intelligent Signal Processing and Communications Systems |
Subtitle of host publication | "The Decade of Intelligent and Green Signal Processing and Communications", ISPACS 2011 |
DOIs | |
Publication status | Published - 1 Dec 2011 |
Event | 19th IEEE International Symposium on Intelligent Signal Processing and Communication Systems, ISPACS 2011 - Chiang Mai, Thailand Duration: 7 Dec 2011 → 9 Dec 2011 |
Conference
Conference | 19th IEEE International Symposium on Intelligent Signal Processing and Communication Systems, ISPACS 2011 |
---|---|
Country/Territory | Thailand |
City | Chiang Mai |
Period | 7/12/11 → 9/12/11 |
Keywords
- face recognition
- Gabor feature
- generalized canonical correlation analysis
- hallucination
- linear regression
ASJC Scopus subject areas
- Computer Networks and Communications
- Signal Processing