La La LiDAR: Large-Scale Layout Generation from LiDAR Data

TL;DR

La La LiDAR is a novel framework for controllable, large-scale LiDAR scene generation that combines semantic scene graphs with diffusion models, enabling customizable and consistent 3D scene synthesis for autonomous driving applications.

Contribution

It introduces a new layout-guided generative framework with semantic scene graph diffusion and relation-aware conditioning, along with large-scale datasets and evaluation metrics for LiDAR scene synthesis.

Findings

Achieves state-of-the-art LiDAR generation quality.

Ensures semantic and spatial consistency in generated scenes.

Improves downstream perception task performance.

Abstract

Controllable generation of realistic LiDAR scenes is crucial for applications such as autonomous driving and robotics. While recent diffusion-based models achieve high-fidelity LiDAR generation, they lack explicit control over foreground objects and spatial relationships, limiting their usefulness for scenario simulation and safety validation. To address these limitations, we propose Large-scale Layout-guided LiDAR generation model ("La La LiDAR"), a novel layout-guided generative framework that introduces semantic-enhanced scene graph diffusion with relation-aware contextual conditioning for structured LiDAR layout generation, followed by foreground-aware control injection for complete scene generation. This enables customizable control over object placement while ensuring spatial and semantic consistency. To support our structured LiDAR generation, we introduce Waymo-SG and nuScenes-SG, two large-scale LiDAR scene graph datasets, along with new evaluation metrics for layout synthesis. Extensive experiments demonstrate that La La LiDAR achieves state-of-the-art performance in both LiDAR generation and downstream perception tasks, establishing a new benchmark for controllable 3D scene generation.