TY  - JOUR
PY  - 2009//
TI  - A fast stereo-based system for detecting and tracking pedestrians from a moving vehicle
JO  - International journal of robotic research
A1  - Bajracharya, Max
A1  - Moghaddam, Baback
A1  - Howard, Andrew William
A1  - Brennan, Shane
A1  - Matthies, Larry H.
SP  - 1466
EP  - 1485
VL  - 28
IS  - 11-12
N2  - In this paper we describe a fully integrated system for detecting, lo- calizing, and tracking pedestrians from a moving vehicle. The system can reliably detect upright pedestrians to a range of 40 m in lightly cluttered urban environments. The system uses range data from stereo vision to segment the scene into regions of interest, from which shape features are extracted and used to classify pedestrians. The regions are tracked using shape and appearance features. Tracking is used to temporally filter classifications to improve performance and to esti- mate the velocity of pedestrians for use in path planning. The end-to- end system runs at 5 Hz on 1024 768 imagery using as tandard 2.4 GHz Intel Core 2 Quad processor, and has been integrated and tested on multiple ground vehicles and environments. We show per- formance on a diverse set of datasets with groundtruth in outdoor environments with varying degrees of pedestrian density and clutter. In highly cluttered urban environments, the detection rates are on a par with state-of-the-art but significantly slower systems.   KEY WORDS--pedestrian detection, human detection, stereo, tracking<p /> 
LA  - 
SN  - 0278-3649
UR  - http://dx.doi.org/10.1177/0278364909341884
ID  - ref1
ER  -