DocumentCode :
3256026
Title :
Real-Time Viola-Jones Face Detection in a Web Browser
Author :
Ephraim, Theo ; Himmelman, Tristan ; Siddiqi, Kaleem
Author_Institution :
Centre for Intell. Machines, McGill Univ., Montreal, QC, Canada
fYear :
2009
fDate :
25-27 May 2009
Firstpage :
321
Lastpage :
328
Abstract :
This paper develops optimizations to the Viola-Jones face detection method to make it suitable for use in a Web browser running on a standard laptop or a desktop equipped with a Web cam. In this setting certain assumptions can be made about the number and location of faces to find, and information from a previous frame can help to localize the search. These optimizations lead to an algorithm which performs real-time face detection using a slow scripting language, even on low-end computers. Specifically, we have implemented the algorithm in Adobe Flash (ActionScript 3.0) and our implementation can be deployed via a Web browser without any extra installation. We invite readers to click on http://flashfacedetection.com and test it for themselves.
Keywords :
face recognition; online front-ends; ActionScript 3.0; Adobe Flash; Viola-Jones face detection; Web browser; Web cam; scripting language; Change detection algorithms; Computer science; Computer vision; Detectors; Face detection; Feeds; Intelligent robots; Object detection; Optimization methods; Robot vision systems; Flash; Haar-like features; face detection; real-time;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer and Robot Vision, 2009. CRV '09. Canadian Conference on
Conference_Location :
Kelowna, BC
Print_ISBN :
978-0-7695-3651-4
Type :
conf
DOI :
10.1109/CRV.2009.48
Filename :
5230503
Link To Document :
بازگشت