Imagine the Unseen: Occluded Pedestrian Detection via Adversarial Feature Completion

TL;DR

This paper introduces a novel adversarial feature completion method to improve occluded pedestrian detection by reconstructing occluded features, leading to significant performance gains on multiple datasets.

Contribution

It proposes a new feature completion approach using adversarial learning to handle occlusions in pedestrian detection, which is a novel solution to intra-class variance issues.

Findings

Significant improvements on CityPersons, especially on heavily occluded pedestrians.

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

Outperforms baseline detectors without relying on extra cues.

Abstract

Pedestrian detection has significantly progressed in recent years, thanks to the development of DNNs. However, detection performance at occluded scenes is still far from satisfactory, as occlusion increases the intra-class variance of pedestrians, hindering the model from finding an accurate classification boundary between pedestrians and background clutters. From the perspective of reducing intra-class variance, we propose to complete features for occluded regions so as to align the features of pedestrians across different occlusion patterns. An important premise for feature completion is to locate occluded regions. From our analysis, channel features of different pedestrian proposals only show high correlation values at visible parts and thus feature correlations can be used to model occlusion patterns. In order to narrow down the gap between completed features and real fully visible ones, we propose an adversarial learning method, which completes occluded features with a generator such that they can hardly be distinguished by the discriminator from real fully visible features. We report experimental results on the CityPersons, Caltech and CrowdHuman datasets. On CityPersons, we show significant improvements over five different baseline detectors, especially on the heavy occlusion subset. Furthermore, we show that our proposed method FeatComp++ achieves state-of-the-art results on all the above three datasets without relying on extra cues.