HG-Lane: High-Fidelity Generation of Lane Scenes under Adverse Weather and Lighting Conditions without Re-annotation

2026-03-10

AI Summary

Existing lane detection datasets severely lack representation under extreme conditions such as rain, snow, fog, and low illumination, leading to significant performance degradation of models in real-world adverse scenarios. To address this limitation, this work proposes a high-fidelity lane scene generation framework that synthesizes diverse adverse weather and lighting conditions without requiring manual re-annotation. Using this framework, we construct the first multi-extreme-condition lane detection benchmark dataset comprising 30,000 images. Experimental results demonstrate that training CLRNet on this benchmark improves its mean F1 score by 20.87%, with a remarkable 38.8% gain in F1@50 under snowy conditions, substantially enhancing the model’s generalization across various challenging environments.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Computer Vision: Adversarial Attacks & Robustness
• Natural Language Processing: Generation

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Lane detection is a crucial task in autonomous driving, as it helps ensure the safe operation of vehicles. However, existing datasets such as CULane and TuSimple contain relatively limited data under extreme weather conditions, including rain, snow, and fog. As a result, detection models trained on these datasets often become unreliable in such environments, which may lead to serious safety-critical failures on the road. To address this issue, we propose HG-Lane, a High-fidelity Generation framework for Lane Scenes under adverse weather and lighting conditions without requiring re-annotation. Based on this framework, we further construct a benchmark that includes adverse weather and lighting scenarios, containing 30,000 images. Experimental results demonstrate that our method consistently and significantly improves the performance of existing lane detection networks. For example, using the state-of-the-art CLRNet, the overall mF1 score on our benchmark increases by 20.87 percent. The F1@50 score for the overall, normal, snow, rain, fog, night, and dusk categories increases by 19.75 percent, 8.63 percent, 38.8 percent, 14.96 percent, 26.84 percent, 21.5 percent, and 12.04 percent, respectively. The code and dataset are available at: https://github.com/zdc233/HG-Lane.

Problem

• lane detection
• adverse weather
• autonomous driving
• data scarcity
• lighting conditions

Innovation

• high-fidelity generation
• lane detection
• adverse weather
• data augmentation
• autonomous driving