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

TL;DR

HG-Lane, a High-fidelity Generation framework for Lane Scenes under adverse weather and lighting conditions without requiring re-annotation is proposed, and a benchmark that includes adverse weather and lighting scenarios, containing 30,000 images is constructed.

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.

Figures (8)

• Figure 1: High-fidelity weather and lighting transformations generated by our HG-Lane framework. Lane labels are preserved exactly, while the remaining scene semantics are kept consistent across Normal, Night, Dusk, Snow, Rain, and Fog conditions.
• Figure 2: Overview of the proposed HG-Lane. The input image is first processed into a fused control map combining color-based masks, Canny edges and lane annotations. In Stage-I, a Canny-ControlNet enforces lane geometry during reverse diffusion in latent space. In Stage-II, an InstructPix2Pix ControlNet optionally refines appearance for "night" and "dusk". Finally, the latent is decoded back to pixel space. All components are pretrained; no fine-tuning is required.
• Figure 3: Results of some baselines. The green lines in the figure represent the predicted values, while blue lines represent ground truth.
• Figure 4: Ablation Study. Experiment #1, #2, #3, and #4 demonstrate the generation results of using Canny or InstructPix2Pix individually, as well as the effects of their combined order.
• Figure 5: Quality Analysis of Generation. Comparison of images generated by different frameworks. The figure illustrates the generation results of different frameworks across five categories (dusk, fog, rain, night, and snow).