PSC-Net: Learning Part Spatial Co-occurrence for Occluded Pedestrian Detection

TL;DR

PSC-Net introduces a graph convolutional network to explicitly model part co-occurrence in pedestrian detection, significantly improving occlusion handling without extra annotations.

Contribution

It proposes a novel GCN-based module for learning part spatial co-occurrence without requiring part annotations or visible bounding boxes.

Findings

Achieves state-of-the-art results on CityPersons and Caltech datasets.

Improves detection performance under heavy occlusion conditions.

Does not rely on additional part annotations or visible bounding boxes.

Abstract

Detecting pedestrians, especially under heavy occlusions, is a challenging computer vision problem with numerous real-world applications. This paper introduces a novel approach, termed as PSC-Net, for occluded pedestrian detection. The proposed PSC-Net contains a dedicated module that is designed to explicitly capture both inter and intra-part co-occurrence information of different pedestrian body parts through a Graph Convolutional Network (GCN). Both inter and intra-part co-occurrence information contribute towards improving the feature representation for handling varying level of occlusions, ranging from partial to severe occlusions. Our PSC-Net exploits the topological structure of pedestrian and does not require part-based annotations or additional visible bounding-box (VBB) information to learn part spatial co-occurrence. Comprehensive experiments are performed on two challenging datasets: CityPersons and Caltech datasets. The proposed PSC-Net achieves state-of-the-art detection performance on both. On the heavy occluded (\textbf{HO}) set of CityPerosns test set, our PSC-Net obtains an absolute gain of 4.0\% in terms of log-average miss rate over the state-of-the-art with same backbone, input scale and without using additional VBB supervision. Further, PSC-Net improves the state-of-the-art from 37.9 to 34.8 in terms of log-average miss rate on Caltech (\textbf{HO}) test set.