Improved shadow removal for robust person tracking in surveillance scenarios

Sanin, Andres, Sanderson, Conrad and Lovell, Brian (2010). Improved shadow removal for robust person tracking in surveillance scenarios. In: Proceedings. 2010 20th International Conference on Pattern Recognition (ICPR 2010). International Conference on Pattern Recognition (20th, ICPR 2010), Istanbul, Turkey, (141-144). 23-26 August 2010. doi:10.1109/ICPR.2010.43

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads

Author Sanin, Andres
Sanderson, Conrad
Lovell, Brian
Title of paper Improved shadow removal for robust person tracking in surveillance scenarios
Conference name International Conference on Pattern Recognition (20th, ICPR 2010)
Conference location Istanbul, Turkey
Conference dates 23-26 August 2010
Proceedings title Proceedings. 2010 20th International Conference on Pattern Recognition (ICPR 2010)   Check publisher's open access policy
Journal name Proceedings - International Conference on Pattern Recognition   Check publisher's open access policy
Place of Publication Los Alamitos, CA, United States
Publisher IEEE
Publication Year 2010
Sub-type Fully published paper
DOI 10.1109/ICPR.2010.43
ISBN 9781424475421
1424475422
ISSN 1051-4651
Start page 141
End page 144
Total pages 4
Collection year 2011
Language eng
Abstract/Summary Shadow detection and removal is an important step employed after foreground detection, in order to improve the segmentation of objects for tracking. Methods reported in the literature typically have a significant trade-off between the shadow detection rate (classifying true shadow areas as shadows) and the shadow discrimination rate (discrimination between shadows and foreground). We propose a method that is able to achieve good performance in both cases, leading to improved tracking in surveillance scenarios. Chromacity information is first used to create a mask of candidate shadow pixels, followed by employing gradient information to remove foreground pixels that were incorrectly included in the mask. Experiments on the CAVIAR dataset indicate that the proposed method leads to considerable improvements in multiple object tracking precision and accuracy. © 2010 IEEE
Keyword Chromacity
Data sets
Discrimination rates
Foreground detection
Gradient informations
Multiple object tracking
Person tracking
Shadow detections
Shadow removal
Q-Index Code E1
Q-Index Status Confirmed Code
Institutional Status UQ
Additional Notes Session MoBT1 Tracking and Surveillance - I. Proceedings series title: International Conference on Pattern Recognition.

 
Versions
Version Filter Type
Citation counts: Scopus Citation Count Cited 42 times in Scopus Article | Citations
Google Scholar Search Google Scholar
Created: Fri, 26 Nov 2010, 18:12:11 EST by Conrad Sanderson on behalf of School of Information Technol and Elec Engineering