Semantic Head Enhanced Pedestrian Detection in a Crowd

TL;DR

This paper introduces HBAN, a pedestrian detection model that leverages semantic head detection and head-body alignment to improve detection accuracy in crowded scenes with occlusion.

Contribution

The paper proposes a novel head-body alignment network that uses weakly labeled semantic head regions and alignment loss to enhance pedestrian detection.

Findings

HBAN outperforms baseline models on CityPersons dataset.

Semantic head detection improves robustness against occlusion.

Head-body alignment contributes significantly to detection accuracy.

Abstract

Pedestrian detection in the crowd is a challenging task because of intra-class occlusion. More prior information is needed for the detector to be robust against it. Human head area is naturally a strong cue because of its stable appearance, visibility and relative location to body. Inspired by it, we adopt an extra branch to conduct semantic head detection in parallel with traditional body branch. Instead of manually labeling the head regions, we use weak annotations inferred directly from body boxes, which is named as `semantic head'. In this way, the head detection is formulated into using a special part of labeled box to detect the corresponding part of human body, which surprisingly improves the performance and robustness to occlusion. Moreover, the head-body alignment structure is explicitly explored by introducing Alignment Loss, which functions in a self-supervised manner. Based on these, we propose the head-body alignment net (HBAN) in this work, which aims to enhance pedestrian detection by fully utilizing the human head prior. Comprehensive evaluations are conducted to demonstrate the effectiveness of HBAN on CityPersons dataset.