Automatic Labelling for Low-Light Pedestrian Detection

TL;DR

This paper introduces an automated infrared-RGB labeling pipeline to improve low-light pedestrian detection in RGB images, leveraging infrared detection to generate training labels that outperform ground-truth labels in certain scenarios.

Contribution

The study presents a novel automated labeling method using infrared detection to enhance low-light pedestrian detection in RGB images, addressing the lack of large public datasets.

Findings

Models trained on generated labels outperform those trained on ground-truth labels in 6 out of 9 cases.

The pipeline effectively transfers infrared detections to RGB labels for training.

The approach improves detection performance in low-light conditions.

Abstract

Pedestrian detection in RGB images is a key task in pedestrian safety, as the most common sensor in autonomous vehicles and advanced driver assistance systems is the RGB camera. A challenge in RGB pedestrian detection, that does not appear to have large public datasets, is low-light conditions. As a solution, in this research, we propose an automated infrared-RGB labeling pipeline. The proposed pipeline consists of 1) Infrared detection, where a fine-tuned model for infrared pedestrian detection is used 2) Label transfer process from the infrared detections to their RGB counterparts 3) Training object detection models using the generated labels for low-light RGB pedestrian detection. The research was performed using the KAIST dataset. For the evaluation, object detection models were trained on the generated autolabels and ground truth labels. When compared on a previously unseen image sequence, the results showed that the models trained on generated labels outperformed the ones trained on ground-truth labels in 6 out of 9 cases for the mAP@50 and mAP@50-95 metrics. The source code for this research is available at https://github.com/BouzoulasDimitrios/IR-RGB-Automated-LowLight-Pedestrian-Labeling