Physics-Aware Diffusion for LiDAR Point Cloud Densification

TL;DR

This paper introduces a physics-aware diffusion framework for LiDAR point cloud densification that improves speed and accuracy by incorporating sensor physics and probabilistic refinement.

Contribution

It presents a novel diffusion-based densification method with sensor physics constraints, achieving high-fidelity results efficiently without retraining detectors.

Findings

Achieves dense LiDAR point clouds in 156ms.

Outperforms previous methods on KITTI-360 and nuScenes datasets.

Boosts off-the-shelf 3D detectors without retraining.

Abstract

LiDAR perception is severely limited by the distance-dependent sparsity of distant objects. While diffusion models can recover dense geometry, they suffer from prohibitive latency and physical hallucinations manifesting as ghost points. We propose Scanline-Consistent Range-Aware Diffusion, a framework that treats densification as probabilistic refinement rather than generation. By leveraging Partial Diffusion (SDEdit) on a coarse prior, we achieve high-fidelity results in just 156ms. Our novel Ray-Consistency loss and Negative Ray Augmentation enforce sensor physics to suppress artifacts. Our method achieves state-of-the-art results on KITTI-360 and nuScenes, directly boosting off-the-shelf 3D detectors without retraining. Code will be made available.