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

arXiv cs.CV / 3/12/2026

📰 NewsTools & Practical UsageModels & Research

共有:

Key Points

HG-Lane introduces a high-fidelity generation framework for lane scenes under adverse weather and lighting conditions that does not require re-annotation, addressing data scarcity in existing datasets.
The authors construct a 30,000-image benchmark with adverse weather and lighting scenarios to evaluate lane detection under challenging conditions.
The framework significantly improves lane detection performance: for CLRNet, overall mF1 on the benchmark rises by 20.87%, and F1@50 scores improve across categories including overall (19.75%), normal (8.63%), snow (38.8%), rain (14.96%), fog (26.84%), night (21.5%), and dusk (12.04%).
They release code and dataset at GitHub, enabling practitioners to reproduce and build upon the work.
This work advances robustness of autonomous driving systems in harsh weather, contributing to safer highway operation and research into data-efficient annotation strategies.

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.