English

V2F-Net: Explicit Decomposition of Occluded Pedestrian Detection

Computer Vision and Pattern Recognition 2021-04-08 v1

Abstract

Occlusion is very challenging in pedestrian detection. In this paper, we propose a simple yet effective method named V2F-Net, which explicitly decomposes occluded pedestrian detection into visible region detection and full body estimation. V2F-Net consists of two sub-networks: Visible region Detection Network (VDN) and Full body Estimation Network (FEN). VDN tries to localize visible regions and FEN estimates full-body box on the basis of the visible box. Moreover, to further improve the estimation of full body, we propose a novel Embedding-based Part-aware Module (EPM). By supervising the visibility for each part, the network is encouraged to extract features with essential part information. We experimentally show the effectiveness of V2F-Net by conducting several experiments on two challenging datasets. V2F-Net achieves 5.85% AP gains on CrowdHuman and 2.24% MR-2 improvements on CityPersons compared to FPN baseline. Besides, the consistent gain on both one-stage and two-stage detector validates the generalizability of our method.

Keywords

Cite

@article{arxiv.2104.03106,
  title  = {V2F-Net: Explicit Decomposition of Occluded Pedestrian Detection},
  author = {Mingyang Shang and Dawei Xiang and Zhicheng Wang and Erjin Zhou},
  journal= {arXiv preprint arXiv:2104.03106},
  year   = {2021}
}

Comments

11 pages, 4 figures

R2 v1 2026-06-24T00:55:21.413Z