Repulsion Loss: Detecting Pedestrians in a Crowd

Detecting individual pedestrians in a crowd remains a challenging problem since the pedestrians often gather together and occlude each other in real-world scenarios. In this paper, we first explore how a state-of-the-art pedestrian detector is harmed by crowd occlusion via experimentation, providing...

Full description

Saved in:
Bibliographic Details
Published in2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition pp. 7774 - 7783
Main Authors Wang, Xinlong, Xiao, Tete, Jiang, Yuning, Shao, Shuai, Sun, Jian, Shen, Chunhua
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.06.2018
Subjects
Benchmark testing
Detectors
Euclidean distance
Feature extraction
Object detection
Proposals
Online AccessGet full text

Cover

More Information
Summary:Detecting individual pedestrians in a crowd remains a challenging problem since the pedestrians often gather together and occlude each other in real-world scenarios. In this paper, we first explore how a state-of-the-art pedestrian detector is harmed by crowd occlusion via experimentation, providing insights into the crowd occlusion problem. Then, we propose a novel bounding box regression loss specifically designed for crowd scenes, termed repulsion loss. This loss is driven by two motivations: the attraction by target, and the repulsion by other surrounding objects. The repulsion term prevents the proposal from shifting to surrounding objects thus leading to more crowd-robust localization. Our detector trained by repulsion loss outperforms the state-of-the-art methods with a significant improvement in occlusion cases.
ISSN:1063-6919
DOI:10.1109/CVPR.2018.00811