# CloudDiffusion: Diffusion-Based Scene Completion in the Point Cloud Domain

arXiv: https://arxiv.org/abs/2606.16048 · HTML: https://mlsanity.com/paper/2606.16048/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in World modeling from LiDAR generation, one of 11 directions in World models.
- Builds on 27 of its 29 references that are in the map.
- First posted on arXiv 14 Jun 2026.

## Placed in

- World models / [World modeling from LiDAR generation](https://mlsanity.com/wm/world-modeling-from-lidar-generation/)
- Diffusion models / [3D, point cloud and scene generation](https://mlsanity.com/diffusion/3d-point-cloud-and-scene-generation/)

## Builds on (most cited first, 27 of 27)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation](https://mlsanity.com/paper/1612.00593/) (2016)
- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [Improved Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2102.09672/) (2021)
- [Perceiver: General Perception with Iterative Attention](https://mlsanity.com/paper/2103.03206/) (2021)
- [Point Transformer V3: Simpler, Faster, Stronger](https://mlsanity.com/paper/2312.10035/) (2023)
- [3D Shape Generation and Completion through Point-Voxel Diffusion](https://mlsanity.com/paper/2104.03670/) (2021)
- [3DShape2VecSet: A 3D Shape Representation for Neural Fields and Generative Diffusion Models](https://mlsanity.com/paper/2301.11445/) (2023)
- [VoxFormer: Sparse Voxel Transformer for Camera-based 3D Semantic Scene Completion](https://mlsanity.com/paper/2302.12251/) (2023)
- [LMSCNet: Lightweight Multiscale 3D Semantic Completion](https://mlsanity.com/paper/2008.10559/) (2020)
- [Sonata: Self-Supervised Learning of Reliable Point Representations](https://mlsanity.com/paper/2503.16429/) (2025)
- [Towards Realistic Scene Generation with LiDAR Diffusion Models](https://mlsanity.com/paper/2404.00815/) (2024)
- [LiDAR Data Synthesis with Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2309.09256/) (2023)
- [Context and Geometry Aware Voxel Transformer for Semantic Scene Completion](https://mlsanity.com/paper/2405.13675/) (2024)
- [Camera-based 3D Semantic Scene Completion with Sparse Guidance Network](https://mlsanity.com/paper/2312.05752/) (2023)
- [Scaling Diffusion Models to Real-World 3D LiDAR Scene Completion](https://mlsanity.com/paper/2403.13470/) (2024)
- [DiffComplete: Diffusion-based Generative 3D Shape Completion](https://mlsanity.com/paper/2306.16329/) (2023)
- [DepthSSC: Monocular 3D Semantic Scene Completion via Depth-Spatial Alignment and Voxel Adaptation](https://mlsanity.com/paper/2311.17084/) (2023)
- [LODE: Locally Conditioned Eikonal Implicit Scene Completion from Sparse LiDAR](https://mlsanity.com/paper/2302.14052/) (2023)
- [Distilling Diffusion Models to Efficient 3D LiDAR Scene Completion](https://mlsanity.com/paper/2412.03515/) (2024)
- [LiDPM: Rethinking Point Diffusion for Lidar Scene Completion](https://mlsanity.com/paper/2504.17791/) (2025)
- [Diffusion Distillation With Direct Preference Optimization For Efficient 3D LiDAR Scene Completion](https://mlsanity.com/paper/2504.11447/) (2025)
- [DiffusionRL: Efficient Training of Diffusion Policies for Robotic Grasping Using RL-Adapted Large-Scale Datasets](https://mlsanity.com/paper/2505.18876/) (2025)
- [EmbodiedDiffusion: End-to-End Traversability-Guided Visual Diffusion for Heterogeneous Robot Navigation](https://mlsanity.com/paper/2512.02851/) (2025)
- [LiNeXt: Revisiting LiDAR Completion with Efficient Non-Diffusion Architectures](https://mlsanity.com/paper/2511.10209/) (2025)
- [ImpedanceDiffusion: Diffusion-Based Global Path Planning for UAV Swarm Navigation with Generative Impedance Control](https://mlsanity.com/paper/2603.09031/) (2026)
- [DiffusionAnything: End-to-End In-context Diffusion Learning for Unified Navigation and Pre-Grasp Motion](https://mlsanity.com/paper/2603.26322/) (2026)

## Abstract (arXiv)

Reconstructing dense 3D scenes from sparse LiDAR point clouds (LiDAR scene completion) is a fundamental challenge in autonomous driving, where diffusion models offer a promising solution. However, existing approaches rely on object-level autoencoders that collapse into unstable global representations at outdoor scale, and suffer from ground truth data corrupted by odometry drift that systematically degrades supervision quality. Furthermore, multi-step diffusion inference incurs prohibitive latency for real-time deployment. We present CloudDiffusion, addressing these issues with three independent components. First, a multi-token Gaussian VAE with cross-attention pooling provides stable scene-scale LiDAR compression as a standalone reconstruction module, avoiding the global-pooling and codebook-collapse failure modes of prior point-cloud autoencoders. Second, an anchor-based ICP ground truth refinement pipeline eliminates drift-induced noise from training supervision, reducing our single-step x0 diffusion teacher's squared Chamfer distance by approximately 16x on SemanticKITTI seq. 08 (0.396 to 0.024 m^2) with no model change (partly aided by the denser, more compact refined references). Third, the same teacher completes scenes in a single x0 step, operating directly in coordinate space, not in the VAE latent. It runs in near real time at 209ms/frame, 65-138x lower inference latency than iterative diffusion baselines. Our results indicate that data quality dominates model design in this regime, and suggest that multi-token latent spaces could serve as a stable first stage for future latent diffusion-based scene completion.