PointDiffusion: Diffusion-Based Scene Completion in the Point Cloud Domain

• URL: http://arxiv.org/abs/2606.16048v1
• Date: Sun, 14 Jun 2026 22:36:34 GMT
• Title: PointDiffusion: Diffusion-Based Scene Completion in the Point Cloud Domain
• Abstract summary: Reconstructing dense 3D scenes from sparse LiDAR point clouds is a fundamental challenge in autonomous driving.<n>Existing approaches rely on object-level autoencoders that collapse into unstable global representations at outdoor scale.<n>We propose a novel multi-token Gaussian VAE with cross-attention pooling for stable scene-scale LiDAR compression.
• Score: 1.7511853008573073
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Reconstructing dense 3D scenes from sparse LiDAR point clouds is a fundamental challenge in autonomous driving, where latent 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 propose a novel multi-token Gaussian VAE with cross-attention pooling for stable scene-scale LiDAR compression, combined with an anchor-based ICP ground truth refinement pipeline that eliminates drift-induced noise from training supervision. Together, these components enable a scaffold-free single-step diffusion completion model that achieves an approximately 16x reduction in squared Chamfer distance on SemanticKITTI seq. 08 (0.396 m^2 to 0.024 m^2), surpasses LiDiff and ScoreLiDAR by 17-19% and 10-11%, respectively, and operates at 25-143x lower inference latency. Our results demonstrate that data quality dominates model design in this regime and that multi-token latent spaces provide a stable first stage for latent diffusion-based scene completion.

Related papers

• PeLAP-A: Adaptive Latent Pruning for Lightweight Latent Diffusion Models [0.0]
  We introduce PeLAP-A (Adaptive Latent Pruning for Diffusion), a lightweight framework that augments a standard latent diffusion pipeline with a learnable channel-wise importance predictor.<n>Under aggressive sparsity regularization, the importance predictor drives all latent channels to near-zero yet the denoising UNet achieves lower diffusion loss.<n>These findings constitute an exploratory study of sparsity dynamics in latent diffusion training, and demonstrate that denoising UNets can remain remarkably robust to latent channel suppression.
  arXiv  Detail & Related papers  (2026-06-22T09:33:19Z)
• Spectrally Regularized Latent Flow Matching for Turbulence Generation [0.0]
  We introduce a latent flow matching framework with a spectrally regularized compression stage.<n>On a 2562 DNS dataset at Re_f approx 2250, replacing an MSE-trained VAE with a zone-weighted log-spectral objective raises deep-dissipation retained spectral power.<n>The improved latent representation also yields a substantially better sampling cost-fidelity tradeoff.
  arXiv  Detail & Related papers  (2026-06-10T06:09:51Z)
• Continuous-Time Distribution Matching for Few-Step Diffusion Distillation [57.28746398500951]
  We introduce Continuous-Time Distribution Matching (CDM), migrating the DMD framework from discrete anchoring to continuous optimization.<n>CDM achieves this through two continuous-time designs. First, we replace the fixed discrete schedule with a dynamic continuous schedule of random length.<n>Second, we propose a continuous-time alignment objective that performs active off-trajectory matching on latents extrapolated via the student's velocity field.
  arXiv  Detail & Related papers  (2026-05-07T14:56:39Z)
• ZID-Net: Zero-Inference Diffusion Prior Decoupling Network for Single Image Dehazing [10.634107398163676]
  Single image dehazing is often constrained by a trade-off between restoration quality and computational efficiency.<n>ZID-Net is a novel framework that explicitly decouples diffusion supervision from feed-forward inference.<n>ZID-Net achieves 40.75 dB PSNR on the synthetic RESIDE dataset and outperforms existing methods with a 1.13 dB gain on real-world datasets.
  arXiv  Detail & Related papers  (2026-04-26T13:44:25Z)
• MISTY: High-Throughput Motion Planning via Mixer-based Single-step Drifting [19.16655963423315]
  MISTY is a generative motion planner that achieves state-of-the-art closed-loopational with pure single-step inference.<n>We introduce a latent-space drifting loss that shifts the complex evolution to the training phase.<n> MISTY operates at over 99 FPS with an end-to-end latency of 10.1 ms.
  arXiv  Detail & Related papers  (2026-04-23T09:50:34Z)
• BadCLIP++: Stealthy and Persistent Backdoors in Multimodal Contrastive Learning [73.46118996284888]
  Research on backdoor attacks against multimodal contrastive learning models faces two key challenges: stealthiness and persistence.<n>We propose BadCLIP++, a unified framework that tackles both challenges.<n>For stealthiness, we introduce a semantic-fusion QR micro-trigger that embeds imperceptible patterns near task-relevant regions.<n>For persistence, we stabilize trigger embeddings via radius shrinkage and centroid alignment.
  arXiv  Detail & Related papers  (2026-02-19T08:31:16Z)
• LiNeXt: Revisiting LiDAR Completion with Efficient Non-Diffusion Architectures [52.59993476989728]
  LiNeXt is a lightweight, non-diffusion network optimized for rapid and accurate point cloud completion.<n>LiNeXt achieves a 199.8x speedup in inference, reduces Chamfer Distance by 50.7%, and uses only 6.1% of the parameters compared with LiDiff.
  arXiv  Detail & Related papers  (2025-11-13T11:28:18Z)
• DiffusionNFT: Online Diffusion Reinforcement with Forward Process [99.94852379720153]
  Diffusion Negative-aware FineTuning (DiffusionNFT) is a new online RL paradigm that optimize diffusion models directly on the forward process via flow matching.<n>DiffusionNFT is up to $25times$ more efficient than FlowGRPO in head-to-head comparisons, while being CFG-free.
  arXiv  Detail & Related papers  (2025-09-19T16:09:33Z)
• Improving Vector-Quantized Image Modeling with Latent Consistency-Matching Diffusion [55.185588994883226]
  We introduce VQ-LCMD, a continuous-space latent diffusion framework within the embedding space that stabilizes training.<n>VQ-LCMD uses a novel training objective combining the joint embedding-diffusion variational lower bound with a consistency-matching (CM) loss.<n>Experiments show that the proposed VQ-LCMD yields superior results on FFHQ, LSUN Churches, and LSUN Bedrooms compared to discrete-state latent diffusion models.
  arXiv  Detail & Related papers  (2024-10-18T09:12:33Z)
• CCD-3DR: Consistent Conditioning in Diffusion for Single-Image 3D Reconstruction [81.98244738773766]
  We present CCD-3DR, which exploits a novel centered diffusion probabilistic model for consistent local feature conditioning. CCD-3DR outperforms all competitors by a large margin, with over 40% improvement.
  arXiv  Detail & Related papers  (2023-08-15T15:27:42Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.