GSRAIN: Physically Calibrated High-/Low-Frequency Rainfall Synthesis for 3D Gaussian Driving Scenes
2026-08-03 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors developed GSRAIN, a new way to add realistic rain effects to 3D driving scenes used to test self-driving cars. They combined detailed raindrop models with a special method that creates rainy looks based on the scene's shape. This approach lets users control how heavy the rain is and keeps the rain consistent from different viewpoints. Their method was better at creating realistic rain than some previous techniques and helped show how rain affects the performance of detection and driving algorithms. Overall, the authors' work helps create more reliable and controllable rain scenes for testing autonomous vehicles.
3D Gaussian SplattingRainfall simulationAutonomous drivingFréchet Inception DistanceDiffusion modelMulti-view consistencyRainfall intensity controlObject detectionClosed-loop drivingSynthetic weather effects
Authors
Fanyu Wang, Longgao Zhang, Junyi Chen
Abstract
Existing rainfall simulation methods for autonomous driving remain limited in physical controllability and multi-view consistency. This paper presents GSRAIN, a high-/low-frequency rainfall synthesis method for 3D Gaussian Splatting (3DGS) driving scenes. GSRAIN constructs a high-frequency raindrop model from measured rainfall data and generates low-frequency rainy appearance using a geometry-aware single-step diffusion model. The two effects are then fused in a unified 3DGS scene, enabling rainfall-intensity control over the range of 0--13~mm/h. The proposed method achieves a Fréchet Inception Distance (FID) of 149.09, outperforming CycleGAN-Turbo (155.71) and WeatherEdit (157.94). Object-detection and closed-loop driving experiments further show that the generated scenes expose scene-dependent performance changes of the evaluated algorithms under controllable rainfall. These results indicate that GSRAIN provides an effective approach for constructing physically controllable, repeatable, and closed-loop-compatible rainy-weather test scenes for autonomous driving.