L3DR: 3D-aware LiDAR Diffusion and Rectification

2026-02-22

AI Summary

Existing range-view–based LiDAR diffusion models often generate artifacts such as depth bleeding and wavy surfaces due to the neglect of 3D geometric constraints. This work proposes a 3D-aware diffusion and correction framework that introduces, for the first time, a 3D residual regression mechanism to explicitly refine geometric structures in 3D space by predicting point-wise 3D offsets. The method incorporates Welsch loss to emphasize informative regions, enabling robust optimization while maintaining low computational overhead and significantly enhancing boundary sharpness and geometric fidelity. The framework is highly generalizable, compatible with various LiDAR diffusion models, and achieves state-of-the-art generation performance across multiple benchmarks, including KITTI, KITTI-360, nuScenes, and Waymo.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Range-view (RV) based LiDAR diffusion has recently made huge strides towards 2D photo-realism. However, it neglects 3D geometry realism and often generates various RV artifacts such as depth bleeding and wavy surfaces. We design L3DR, a 3D-aware LiDAR Diffusion and Rectification framework that can regress and cancel RV artifacts in 3D space and restore local geometry accurately. Our theoretical and empirical analysis reveals that 3D models are inherently superior to 2D models in generating sharp and authentic boundaries. Leveraging such analysis, we design a 3D residual regression network that rectifies RV artifacts and achieves superb geometry realism by predicting point-level offsets in 3D space. On top of that, we design a Welsch Loss that helps focus on local geometry and ignore anomalous regions effectively. Extensive experiments over multiple benchmarks including KITTI, KITTI360, nuScenes and Waymo show that the proposed L3DR achieves state-of-the-art generation and superior geometry-realism consistently. In addition, L3DR is generally applicable to different LiDAR diffusion models with little computational overhead.

Problem

• LiDAR diffusion
• 3D geometry realism
• range-view artifacts
• depth bleeding
• wavy surfaces

Innovation

• 3D-aware LiDAR generation
• diffusion rectification
• geometry realism
• range-view artifacts
• point-level offset regression