Abstract
Our goal is to build an automated system for face recognition. Such a system for a realistic application is likely to have thousands, possibly millions of faces. Hence, it is essential to have a compact representation for a face. So an important issue is the minimum spatial and grayscale resolutions necessary for a pattern to be detected as a face and then identified. Several experiments were performed to estimate these limits using a collection of 64 faces imaged under very different conditions. All experiments were performed using human observers. The results indicate that there is enough information in 32 × 32 × 4 bpp images for human eyes to detect and identify the faces. Thus an automated system could represent a face using only 512 bytes.
Original language | English (US) |
---|---|
Title of host publication | Proceedings of SPIE - The International Society for Optical Engineering |
Editors | Bernice E. Rogowitz, Michael H. Brill, Jan P. Allebach |
Publisher | Publ by Int Soc for Optical Engineering |
Pages | 81-89 |
Number of pages | 9 |
Volume | 1453 |
State | Published - 1991 |
Event | SPIE/IS&T Symposium on Electronic Imaging Science and Technology - San Jose, CA, USA Duration: Feb 24 1991 → Mar 1 1991 |
Other
Other | SPIE/IS&T Symposium on Electronic Imaging Science and Technology |
---|---|
City | San Jose, CA, USA |
Period | 2/24/91 → 3/1/91 |
ASJC Scopus subject areas
- Electrical and Electronic Engineering
- Condensed Matter Physics