V2F-Net: Explicit Decomposition of Occluded Pedestrian Detection

TL;DR

V2F-Net improves occluded pedestrian detection by explicitly decomposing the task into visible region detection and full body estimation, utilizing a novel part-aware module for enhanced accuracy.

Contribution

The paper introduces V2F-Net, a new framework with two sub-networks and a part-aware module, to better handle occlusion in pedestrian detection tasks.

Findings

Achieves 5.85% AP improvement on CrowdHuman dataset.

Achieves 2.24% MR-2 reduction on CityPersons dataset.

Demonstrates effectiveness across different detector architectures.

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.