DocumentCode :
3221768
Title :
Selecting and evaluating data for training a pedestrian detector for crowded conditions
Author :
Simonnet, Damien ; Velastin, Sergio A. ; Orwell, James ; Turkbeyler, Esin
Author_Institution :
Digital Imaging Res. Centre, Kingston Univ., Kingston upon Thames, UK
fYear :
2011
fDate :
16-18 Nov. 2011
Firstpage :
174
Lastpage :
179
Abstract :
Computer vision algorithms for pedestrian detection are often based on classification derived from supervised learning and therefore require training data, which can be built by using generic or specific images. In this field, INRIA datasets are a standard reference but include only few CCTV camera samples. Therefore, for a CCTV camera system it might be interesting to have specific training data. However, in practice it is impossible to create a training data for each camera view. Thus, this paper presents an evaluation of a pedestrian detection algorithm in crowded conditions in relation to the training data, and shows that a CCTV camera training data provides better results and can be reused for similar CCTV camera views.
Keywords :
closed circuit television; computer vision; image sensors; object detection; video surveillance; CCTV camera system; INRIA datasets; computer vision algorithms; crowded conditions; data evaluation; data selection; pedestrian detection algorithm; pedestrian detector; supervised learning; surveillance videos; Cameras; Conferences; Detectors; Feature extraction; Humans; Image edge detection; Training;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Signal and Image Processing Applications (ICSIPA), 2011 IEEE International Conference on
Conference_Location :
Kuala Lumpur
Print_ISBN :
978-1-4577-0243-3
Type :
conf
DOI :
10.1109/ICSIPA.2011.6144127
Filename :
6144127
Link To Document :
بازگشت