A Coulomb Force Inspired Loss Function for High-Performance Pedestrian Detection

Zhe Wang, Jun Wang, Yezhou Yang, Junliang Xing

Research output: Contribution to journalArticlepeer-review

Abstract

Pedestrian detection has received considerable research interest due to its wide application and has made significant progress along with the development of deep neural networks. However, crowd occlusion still remains a significant challenge to current state-of-the-art pedestrian detectors due to the complication in formulating interactions between occluded instances. Inspired by the Coulomb force, we in this work set each proposal as a single electric charge and define the attractive and repulsive forces to model the interaction between ground truths and assigned proposals. This design is driven by two motivations: the attractive force pulls bounding boxes toward their assigned targets, aggregating them compactly around the ground truths. The repulsive force pushes bounding boxes away from other instances, preventing them from shifting to surrounding pedestrians. With this insight, we propose a novel bounding box regression loss and achieve more robust localization performance in crowded scenes without introducing any computational overhead. Extensive experimental evaluations on the CityPersons and CrowdHuman benchmarks demonstrate consistent state-of-the-art performance.

Original languageEnglish (US)
Pages (from-to)2318-2322
Number of pages5
JournalIEEE Signal Processing Letters
Volume29
DOIs
StatePublished - 2022
Externally publishedYes

Keywords

  • Bounding box regression
  • Coulomb force
  • loss function
  • pedestrian detection

ASJC Scopus subject areas

  • Signal Processing
  • Applied Mathematics
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'A Coulomb Force Inspired Loss Function for High-Performance Pedestrian Detection'. Together they form a unique fingerprint.

Cite this