Filtered Feature Channels for Pedestrian Detection

Shanshan Zhang, Rodrigo Benenson, Bernt Schiele

CVPR 2015

/cvpr/2015/zhang2015cvpr-filtered/

Abstract

This paper starts from the observation that multiple top performing pedestrian detectors can be modelled by using an intermediate layer filtering low-level features in combination with a boosted decision forest. Based on this observation we propose a unifying framework and experimentally explore different filter families. We report extensive results enabling a systematic analysis. Using filtered channel features we obtain top performance on the challenging Caltech and KITTI datasets, while using only HOG+LUV as low-level features. When adding optical flow features we further improve detection quality and report the best known result on the Caltech dataset, reaching 93% recall at 1 FPPI.

PDF CVPR Semantic Scholar

Cite

Text

Zhang et al. "Filtered Feature Channels for Pedestrian Detection." Conference on Computer Vision and Pattern Recognition, 2015.

Markdown

[Zhang et al. "Filtered Feature Channels for Pedestrian Detection." Conference on Computer Vision and Pattern Recognition, 2015.](https://mlanthology.org/cvpr/2015/zhang2015cvpr-filtered/)

BibTeX

@inproceedings{zhang2015cvpr-filtered,
  title     = {{Filtered Feature Channels for Pedestrian Detection}},
  author    = {Zhang, Shanshan and Benenson, Rodrigo and Schiele, Bernt},
  booktitle = {Conference on Computer Vision and Pattern Recognition},
  year      = {2015},
  url       = {https://mlanthology.org/cvpr/2015/zhang2015cvpr-filtered/}
}