Boosted Convolutional Neural Networks (BCNN) for Pedestrian Detection

Abstract

A boosted convolutional neural network (BCNN) system is proposed to enhance the pedestrian detection performance in this work. Being inspired by the classic boosting idea, we develop a weighted loss function that emphasizes challenging samples in training a convolutional neural network (CNN). Two types of samples are considered challenging: 1) samples with detection scores falling in the decision boundary, and 2) temporally associated samples with inconsistent scores. A weighting scheme is designed for each of them. Finally, we train a boosted fusion layer to benefit from the integration of these two weighting schemes. We use the Fast-RCNN as the baseline, and test the corresponding BCNN on the Caltech pedestrian dataset in the experiment, and show a significant performance gain of the BCNN over its baseline.

Cite

Text

Wu et al. "Boosted Convolutional Neural Networks (BCNN) for Pedestrian Detection." IEEE/CVF Winter Conference on Applications of Computer Vision, 2017. doi:10.1109/WACV.2017.66

Markdown

[Wu et al. "Boosted Convolutional Neural Networks (BCNN) for Pedestrian Detection." IEEE/CVF Winter Conference on Applications of Computer Vision, 2017.](https://mlanthology.org/wacv/2017/wu2017wacv-boosted/) doi:10.1109/WACV.2017.66

BibTeX

@inproceedings{wu2017wacv-boosted,
  title     = {{Boosted Convolutional Neural Networks (BCNN) for Pedestrian Detection}},
  author    = {Wu, Chi-Hao and Gan, Weihao and Lan, De and Kuo, C.-C. Jay},
  booktitle = {IEEE/CVF Winter Conference on Applications of Computer Vision},
  year      = {2017},
  pages     = {540-549},
  doi       = {10.1109/WACV.2017.66},
  url       = {https://mlanthology.org/wacv/2017/wu2017wacv-boosted/}
}