A Fast Stereo-based System for Detecting and Tracking Pedestrians from a Moving Vehicle

被引:46
作者
Bajracharya, Max [1 ]
Moghaddam, Baback [1 ]
Howard, Andrew [1 ]
Brennan, Shane [1 ]
Matthies, Larry H. [1 ]
机构
[1] CALTECH, Jet Prop Lab, Pasadena, CA 91109 USA
关键词
pedestrian detection; human detection; stereo; tracking;
D O I
10.1177/0278364909341884
中图分类号
TP24 [机器人技术];
学科分类号
080202 ; 1405 ;
摘要
In this paper we describe a fully integrated system for detecting, localizing, 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 estimate the velocity of pedestrians for use in path planning. The end-to-end system runs at 5 Hz on 1,024 x 768 imagery using a standard 2.4 GHz Intel Core 2 Quad processor, and has been integrated and tested on multiple ground vehicles and environments. We show performance 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.
引用
收藏
页码:1466 / 1485
页数:20
相关论文
共 28 条
[1]  
ABDALMAGEED W, 2007, IEEE INT TRANSP SYST
[2]  
[Anonymous], 2003, IEEE INT C COMP VIS
[3]  
[Anonymous], 2008, IEEE C COMP VIS PATT
[4]  
[Anonymous], 2007, IEEE C COMP VIS PATT
[5]  
[Anonymous], IEEE INT VEH S
[6]  
[Anonymous], IEEE C COMP VIS PATT
[7]  
[Anonymous], IEEE C COMP VIS PATT
[8]  
ARNDT R, 2007, IEEE INT VEH S
[9]  
BAJRACHARYA M, 2008, P SPIE, V6962
[10]  
Bertozzi M., 2007, IEEE INT TRANSP SYST