FogGuard: guarding YOLO against fog using perceptual loss

TL;DR

FogGuard is a fog-aware object detection network that enhances YOLO's performance in foggy conditions using a novel perceptual loss, significantly improving detection accuracy without increasing inference time.

Contribution

We introduce FogGuard, which incorporates a Teacher-Student Perceptual loss to improve YOLO's robustness in foggy environments, a novel approach compared to existing enhancement and domain adaptation methods.

Findings

Achieves 69.43% mAP on RTTS, outperforming YOLOv3's 57.78%.

Improves detection robustness in foggy conditions with minimal inference overhead.

Demonstrates effectiveness across standard datasets like PASCAL VOC and RTTS.

Abstract

In this paper, we present FogGuard, a novel fog-aware object detection network designed to address the challenges posed by foggy weather conditions. Autonomous driving systems heavily rely on accurate object detection algorithms, but adverse weather conditions can significantly impact the reliability of deep neural networks (DNNs). Existing approaches include image enhancement techniques like IA-YOLO and domain adaptation methods. While image enhancement aims to generate clear images from foggy ones, which is more challenging than object detection in foggy images, domain adaptation does not require labeled data in the target domain. Our approach involves fine-tuning on a specific dataset to address these challenges efficiently. FogGuard compensates for foggy conditions in the scene, ensuring robust performance by incorporating YOLOv3 as the baseline algorithm and introducing a unique Teacher-Student Perceptual loss for accurate object detection in foggy environments. Through comprehensive evaluations on standard datasets like PASCAL VOC and RTTS, our network significantly improves performance, achieving a 69.43\% mAP compared to YOLOv3's 57.78\% on the RTTS dataset. Additionally, we demonstrate that while our training method slightly increases time complexity, it doesn't add overhead during inference compared to the regular YOLO network.