TLDR: Traffic Light Detection using Fourier Domain Adaptation in Hostile WeatheR

2024-11-12

arXiv.org

AI Summary

To address the severe performance degradation of traffic light detection under adverse weather conditions and the scarcity of annotated data, this paper proposes a cross-weather generalization method integrating Fourier Domain Adaptation (FDA) and semi-supervised learning. First, we construct a class-balanced source domain by fusing the LISA and S2TLD datasets and synthesize rain/fog-corrupted images to form the target domain. Second, we pioneer the application of FDA to traffic light detection, enabling efficient frequency-domain alignment between source and target domains. Third, we embed this adaptation into the YOLOv8 framework with a semi-supervised training strategy to maximize annotation efficiency. Experiments demonstrate significant improvements: +9.51% in mAP₅₀, +19.50% in mAP₅₀–₉₅, +5.19% in Precision, and +14.80% in Recall; the overall model achieves a +23.81% average gain in mAP₅₀–₉₅, markedly enhancing robustness in rainy and foggy scenarios.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Application Domains: Transportation

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

The scarcity of comprehensive datasets in the traffic light detection and recognition domain and the poor performance of state-of-the-art models under hostile weather conditions present significant challenges. To address these issues, this paper proposes a novel approach by merging two widely used datasets, LISA and S2TLD. The merged dataset is further processed to tackle class imbalance, a common problem in this domain. This merged dataset becomes our source domain. Synthetic rain and fog are added to the dataset to create our target domain. We employ Fourier Domain Adaptation (FDA) to create a final dataset with a minimized domain gap between the two datasets, helping the model trained on this final dataset adapt to rainy and foggy weather conditions. Additionally, we explore Semi-Supervised Learning (SSL) techniques to leverage the available data more effectively. Experimental results demonstrate that models trained on FDA-augmented images outperform those trained without FDA across confidence-dependent and independent metrics, like mAP50, mAP50-95, Precision, and Recall. The best-performing model, YOLOv8, achieved a Precision increase of 5.1860%, Recall increase of 14.8009%, mAP50 increase of 9.5074%, and mAP50-95 increase of 19.5035%. On average, percentage increases of 7.6892% in Precision, 19.9069% in Recall, 15.8506% in mAP50, and 23.8099% in mAP50-95 were observed across all models, highlighting the effectiveness of FDA in mitigating the impact of adverse weather conditions on model performance. These improvements pave the way for real-world applications where reliable performance in challenging environmental conditions is critical.

Problem

• Detect traffic lights in adverse weather conditions
• Reduce computational overhead in training and deployment
• Improve model performance under rain and fog

Innovation

• Uses Fourier Domain Adaptation for weather adaptation
• Modifies training data without architectural changes
• Enhances detection metrics significantly in adverse weather