Faculty of Informatics Vienna University of Technology Institute of Computer Aided Automation PRIP Home PRIP Home
Personal tools
You are here: Home Teaching PRIP-Talks Stéphanie Pleyer: Evaluation of person detectors in surveillance videos

Stéphanie Pleyer: Evaluation of person detectors in surveillance videos

When Jul 30, 2009
from 04:15 pm to 04:35 pm
Where Sem 183/2
Add event to calendar vCal
Society's increasing need for security causes a growing use of video surveillance systems in public places. One major scientific aim in video surveillance applications is to be able to detect persons automatically. This project in the domain of machine vision and pattern recognition tends to automatically detect persons in video data from cameras.
This final talk of my internship provides an evaluation of two state of the art person detectors. We evaluate different pre processing steps to improve the reliability of these approaches. Data sets include videos of places of public life, surveillance videos in bank branches and subterranean garage. We aim for an accurate and fast approach to localize persons.
The aim of this project is the comparison of existing state of the art systems for human detection in surveillance camera videos: We use the approach of boosted Haar-like features and compare it with a fast classification of HoG features. Real life cameras videos are used to test the detector. The possibility offered by this technology is to store only the parts of the video data where some suspicious actions occur, and therefore to be able to have far less data to store than now.
Document Actions