Minimum resolution for human face detection and identification

Research output: Chapter in Book/Report/Conference proceedingConference contribution

12 Scopus citations

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 languageEnglish (US)
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
EditorsBernice E. Rogowitz, Michael H. Brill, Jan P. Allebach
PublisherPubl by Int Soc for Optical Engineering
Pages81-89
Number of pages9
Volume1453
StatePublished - 1991
EventSPIE/IS&T Symposium on Electronic Imaging Science and Technology - San Jose, CA, USA
Duration: Feb 24 1991Mar 1 1991

Other

OtherSPIE/IS&T Symposium on Electronic Imaging Science and Technology
CitySan Jose, CA, USA
Period2/24/913/1/91

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Fingerprint Dive into the research topics of 'Minimum resolution for human face detection and identification'. Together they form a unique fingerprint.

Cite this