Pedestrian detection at 100 frames per second

Authors:
Luc Van Gool
Affiliations:
ESAT-PSI-VISICS/IBBT, Katholieke Universiteit Leuven, Belgium
Venue:
CVPR '12 Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Year:
2012

Citing 0
Cited 7

Crosstalk cascades for frame-rate pedestrian detection

ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part II
A three-layered approach to facade parsing

ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part VII
Fast stixel computation for fast pedestrian detection

ECCV'12 Proceedings of the 12th international conference on Computer Vision - Volume Part III
Inserting virtual pedestrians into pedestrian groups video with behavior consistency

The Visual Computer: International Journal of Computer Graphics
Fusion of 3D-LIDAR and camera data for scene parsing

Journal of Visual Communication and Image Representation
Combining patch matching and detection for robust pedestrian tracking in monocular calibrated cameras

Pattern Recognition Letters
Boosting masked dominant orientation templates for efficient object detection

Computer Vision and Image Understanding

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present a new pedestrian detector that improves both in speed and quality over state-of-the-art. By efficiently handling different scales and transferring computation from test time to training time, detection speed is improved. When processing monocular images, our system provides high quality detections at 50 fps. We also propose a new method for exploiting geometric context extracted from stereo images. On a single CPU+GPU desktop machine, we reach 135 fps, when processing street scenes, from rectified input to detections output.