Denoising Diffusion Probabilistic Model for Point Cloud Compression at Low Bit-Rates

Gabriele Spadaro, Alberto Presta, Jhony H. Giraldo, Marco Grangetto, Wei Hu +3 more
5/19/2025
cs.CVcs.AIcs.LG

Official code: EIDOSLAB/DDPM-PCC (4 stars, MIT, last commit 2025-06-30).

Abstract

Efficient compression of low-bit-rate point clouds is critical for bandwidth-constrained applications. However, existing techniques mainly focus on high-fidelity reconstruction, requiring many bits for compression. This paper proposes a "Denoising Diffusion Probabilistic Model" (DDPM) architecture for point cloud compression (DDPM-PCC) at low bit-rates. A PointNet encoder produces the condition vector for the generation, which is then quantized via a learnable vector quantizer. This configuration allows to achieve a low bitrates while preserving quality. Experiments on ShapeNet and ModelNet40 show improved rate-distortion at low rates compared to standardized and state-of-the-art approaches. We publicly released the code at https://github.com/EIDOSLAB/DDPM-PCC.

View on arXivView PDF

Code Implementations(1)

EIDOSLAB/DDPM-PCCOfficial100%
MIT
40C++, CudaApr 1, 20251 years agoMIT

Similar Papers

Diff-PCC: Diffusion-based Neural Compression for 3D Point Clouds

83% match2024

DiffCom: Decoupled Sparse Priors Guided Diffusion Compression for Point Clouds

83% match2024

Lossy Image Compression with Foundation Diffusion Models

82% match1 repo1362024

Diffusion Probabilistic Models for 3D Point Cloud Generation

82% match1 repo7892021

A Residual Diffusion Model for High Perceptual Quality Codec Augmentation

82% match2023

Improving Denoising Diffusion Probabilistic Models via Exploiting Shared Representations

81% match2023

Cite this paper

@article{spadaro2025denoising,
  title  = {Denoising Diffusion Probabilistic Model for Point Cloud Compression at Low Bit-Rates},
  author = {Gabriele Spadaro and Alberto Presta and Jhony H. Giraldo and Marco Grangetto and Wei Hu and Giuseppe Valenzise and Attilio Fiandrotti and Enzo Tartaglione},
  year   = {2025},
  eprint = {2505.13316},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2505.13316v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2505.13316.svg)](https://codeofpaper.com/paper/2505.13316)

Discussion