Operator Attention Based Surveillance System
Surveillance systems based on security cameras are widely used. Surveillance operators synchronously monitor a large number of video streams from many cameras. The operators can overlook some important events due to heavy workload. As a solution to this problem, Res. Asst. Ulas Vural and Assoc. Prof. Yusuf Sinan Akgul from Gebze Institute of
Technology developed a system which can detect the actions overlooked by the surveillance operator. The system tracks the eye gaze positions of the operator to decide if any important actions are missed. A short summary video is automatically built to be shown to the operator for a second try. This work was the topic of the news by the internationally respected science magazine New Scientist with the title "Eyeball spy turns the tables on Big Brother". A number of local and national newspapers and television channels also made news about the work.
Project Supported by
TUBITAK, Project No:110E033, Total Budget: 133,115 TL, 2010 - 2012
Publications
- Ulas Vural and Yusuf Sinan Akgul, Operator interest based surveillance video synthesis, International Conference on Pattern Recognition,,
- ·21st International Conference on Pattern Recognition, Tokyo, 2012
- Ulas Vural and Yusuf Sinan Akgul. "Operator Attention Based Video Surveillance", The Eleventh IEEE International Workshop on Visual Surveillance, Barcelona, 2011.
- Ulas Vural and Yusuf Sinan Akgul, "A Parallel Non-Linear Surveillance Video Synopsis System with Operator Eye-Gaze Input", Video Surveillance, Weiyao Lin (Ed.), ISBN: 978-953-307-436-8, InTech, 2011.
- Ulas Vural and Yusuf Sinan Akgul. "Eye-gaze Based Real-time Surveillance Video Synopsis", the Special Issue of Pattern Recognition Letters Image/Video-based Pattern Analysis and HCI Applications, 2009
Patent Application
- Akgul Y.S., Vural, U., and Yildiz, A.O., (2010). Video
- ·summary system, US Patent App. 12/728,172.
In the News
- New Scientist, 14 April 2009 by Paul Marks, “Eyeball spy turns the tables on Big Brother”