Advanced search
1 file | 871.74 KB

An occlusion-robust feature selection framework in Pedestrian detection

Zhixin Guo (UGent) , Wenzhi Liao (UGent) , Yifan Xiao, Peter Veelaert (UGent) and Wilfried Philips (UGent)
(2018) SENSORS. 18(7).
Author
Organization
Abstract
Better features have been driving the progress of pedestrian detection over the past years. However, as features become richer and higher dimensional, noise and redundancy in the feature sets become bigger problems. These problems slow down learning and can even reduce the performance of the learned model. Current solutions typically exploit dimension reduction techniques. In this paper, we propose a simple but effective feature selection framework for pedestrian detection. Moreover, we introduce occluded pedestrian samples into the training process and combine it with a new feature selection criterion, which enables improved performances for occlusion handling problems. Experimental results on the Caltech Pedestrian dataset demonstrate the efficiency of our method over the state-of-art methods, especially for the occluded pedestrians.
Keywords
pedestrian detection, feature selection, occlusion handling, deep learning

Downloads

  • sensors-18-02272.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 871.74 KB

Citation

Please use this url to cite or link to this publication:

Chicago
Guo, Zhixin, Wenzhi Liao, Yifan Xiao, Peter Veelaert, and Wilfried Philips. 2018. “An Occlusion-robust Feature Selection Framework in Pedestrian Detection.” Sensors 18 (7).
APA
Guo, Zhixin, Liao, W., Xiao, Y., Veelaert, P., & Philips, W. (2018). An occlusion-robust feature selection framework in Pedestrian detection. SENSORS, 18(7).
Vancouver
1.
Guo Z, Liao W, Xiao Y, Veelaert P, Philips W. An occlusion-robust feature selection framework in Pedestrian detection. SENSORS. 2018;18(7).
MLA
Guo, Zhixin, Wenzhi Liao, Yifan Xiao, et al. “An Occlusion-robust Feature Selection Framework in Pedestrian Detection.” SENSORS 18.7 (2018): n. pag. Print.
@article{8568925,
  abstract     = {Better features have been driving the progress of pedestrian detection over the past years.
However, as features become richer and higher dimensional, noise and redundancy in the feature sets
become bigger problems. These problems slow down learning and can even reduce the performance
of the learned model. Current solutions typically exploit dimension reduction techniques. In this
paper, we propose a simple but effective feature selection framework for pedestrian detection.
Moreover, we introduce occluded pedestrian samples into the training process and combine it with
a new feature selection criterion, which enables improved performances for occlusion handling
problems. Experimental results on the Caltech Pedestrian dataset demonstrate the efficiency of our
method over the state-of-art methods, especially for the occluded pedestrians.},
  author       = {Guo, Zhixin and Liao, Wenzhi and Xiao, Yifan and Veelaert, Peter and Philips, Wilfried},
  issn         = {1424-8220},
  journal      = {SENSORS},
  keyword      = {pedestrian detection,feature selection,occlusion handling,deep learning},
  language     = {eng},
  number       = {7},
  pages        = {18},
  title        = {An occlusion-robust feature selection framework in Pedestrian detection},
  url          = {http://dx.doi.org/10.3390/s18072272},
  volume       = {18},
  year         = {2018},
}

Altmetric
View in Altmetric