NMS by Representative Region: Towards Crowded Pedestrian Detection by Proposal Pairing

TL;DR

This paper introduces a novel pedestrian detection method for crowded scenes that uses representative region NMS and a paired-box model to improve detection accuracy amidst occlusions.

Contribution

It proposes a new NMS approach leveraging visible parts and a paired-box model to enhance pedestrian detection in crowded, occluded scenes.

Findings

Effective removal of redundant boxes in crowded scenes

Improved detection accuracy on CrowdHuman and CityPersons benchmarks

Better performance on both full and visible pedestrian detection tasks

Abstract

Although significant progress has been made in pedestrian detection recently, pedestrian detection in crowded scenes is still challenging. The heavy occlusion between pedestrians imposes great challenges to the standard Non-Maximum Suppression (NMS). A relative low threshold of intersection over union (IoU) leads to missing highly overlapped pedestrians, while a higher one brings in plenty of false positives. To avoid such a dilemma, this paper proposes a novel Representative Region NMS approach leveraging the less occluded visible parts, effectively removing the redundant boxes without bringing in many false positives. To acquire the visible parts, a novel Paired-Box Model (PBM) is proposed to simultaneously predict the full and visible boxes of a pedestrian. The full and visible boxes constitute a pair serving as the sample unit of the model, thus guaranteeing a strong correspondence between the two boxes throughout the detection pipeline. Moreover, convenient feature integration of the two boxes is allowed for the better performance on both full and visible pedestrian detection tasks. Experiments on the challenging CrowdHuman and CityPersons benchmarks sufficiently validate the effectiveness of the proposed approach on pedestrian detection in the crowded situation.