URL-Sim: Unified and Regulated LiDAR Simulation for High-Fidelity Reconstruction of Dynamic Driving Scenes
Paper · https://arxiv.org/abs/2611.12932
arXiv:2611.12932v1 Announce Type: new Abstract: Reconstructing high-fidelity LiDAR simulation scenes from real-world driving logs is essential for cost-effective safety validation of autonomous driving systems. While simulators based on Neural Radiance Fields (NeRF) achieve realistic scene reconstruction, their large-scale adoption is limited by slow rendering speeds. Recent 3D Gaussian splatting (3DGS) approaches have improved rendering efficiency through explicit primitive representation and differentiable rasterization. However, existing 3DGS-based methods remain largely restricted to rigid vehicles and produce redundant primitives through unconstrained densification. To address these limitations, we propose URL-Sim, a unified and regulated dynamic LiDAR simulation framework. URL-Sim constructs a dynamic Gaussian scene graph that decouples the modeling of traffic participants, including vehicles, pedestrians, and cyclists, enabling accurate reconstruction of heterogeneous dynamic objects within a single framework. It further introduces a regulated Gaussian densification strategy that ranks primitives with multi-dimensional importance scores and allocates Gaussians to the most critical regions while keeping the total primitive count bounded. Experiments on the Waymo and KITTI-360 datasets demonstrate that URL-Sim achieves higher quantitative accuracy and visual quality than existing baselines under comparable primitive budgets, and accurately recovers the geometry and poses of diverse dynamic traffic participants.
First seen by Oriel on .
AI summary
URL-Sim is a dynamic LiDAR simulation framework that improves reconstruction of heterogeneous driving scenes while bounding the number of 3D Gaussian primitives.
- A dynamic Gaussian scene graph separately models vehicles, pedestrians, and cyclists within one framework.
- Its regulated densification strategy ranks primitives using multi-dimensional importance scores and allocates Gaussians to critical regions.
- The method keeps the total primitive count bounded, addressing redundant primitives created by unconstrained densification.
- On Waymo and KITTI-360, URL-Sim outperforms existing baselines in quantitative accuracy and visual quality under comparable primitive budgets.
- It accurately recovers the geometry and poses of diverse dynamic traffic participants.
#lidar simulation #3d gaussian splatting #autonomous driving #scene reconstruction
Generated by gpt-5.6-sol. Check the original for details.
Aliases
| Source | External ID | URL |
|---|---|---|
| arXiv AI Papers | oai:arXiv.org:2611.12932v1 | https://arxiv.org/abs/2611.12932 Read discussion |