{"slug": "url-sim-unified-and-regulated-lidar-simulation-for-high-fidelity-reconstruction", "title": "URL-Sim: Unified and Regulated LiDAR Simulation for High-Fidelity Reconstruction of Dynamic Driving Scenes", "summary": "Researchers introduced URL-Sim, a unified dynamic LiDAR simulation framework that builds a Gaussian scene graph to separately model vehicles, pedestrians and cyclists, and uses a regulated densification strategy that ranks primitives by multi-dimensional importance scores while keeping the total primitive count bounded. Experiments on the Waymo and KITTI-360 datasets showed URL-Sim achieved higher quantitative accuracy and visual quality than existing baselines under comparable primitive budgets, and accurately recovered the geometry and poses of diverse dynamic traffic participants. The work targets cost-effective safety validation of autonomous driving systems, addressing the slow rendering of NeRF-based simulators and the rigid-vehicle limits of prior 3D Gaussian splatting methods.", "body_md": "arXiv:2611.12932v1 Announce Type: new \nAbstract: 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.", "url": "https://wpnews.pro/news/url-sim-unified-and-regulated-lidar-simulation-for-high-fidelity-reconstruction", "canonical_source": "https://arxiv.org/abs/2611.12932", "published_at": "2026-11-11 05:00:00+00:00", "updated_at": "2026-11-11 05:23:59.571763+00:00", "lang": "en", "topics": ["autonomous-vehicles", "computer-vision", "ai-research", "machine-learning"], "entities": ["URL-Sim", "Waymo", "KITTI-360", "NeRF", "3D Gaussian splatting"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/url-sim-unified-and-regulated-lidar-simulation-for-high-fidelity-reconstruction", "markdown": "https://wpnews.pro/news/url-sim-unified-and-regulated-lidar-simulation-for-high-fidelity-reconstruction.md", "text": "https://wpnews.pro/news/url-sim-unified-and-regulated-lidar-simulation-for-high-fidelity-reconstruction.txt", "jsonld": "https://wpnews.pro/news/url-sim-unified-and-regulated-lidar-simulation-for-high-fidelity-reconstruction.jsonld"}}