Denoising Diffusion Probabilistic Model for Point Cloud Compression at Low Bit-Rates
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.
Authors
- Attilio Fiandrotti (ORCID: https://orcid.org/0000-0002-9991-6822)
- Giuseppe Valenzise (ORCID: https://orcid.org/0000-0002-5840-5743)
- Enzo Tartaglione (ORCID: https://orcid.org/0000-0003-4274-8298)
- Jhony H. Giraldo (ORCID: https://orcid.org/0000-0002-0039-1270)
- Wei Hu (ORCID: https://orcid.org/0000-0002-9860-0922)
- Marco Grangetto (ORCID: https://orcid.org/0000-0002-2709-7864)
- Alberto Presta (ORCID: https://orcid.org/0000-0001-6590-3604)
- Gabriele Spadaro (ORCID: https://orcid.org/0009-0008-4786-1074)
Institutions
- Centre National de la Recherche Scientifique (FR)
- Peking University (CN)
- Laboratoire des signaux et systèmes (FR)
- Institut Polytechnique de Paris (FR)
- Laboratoire Traitement et Communication de l’Information (FR)
- University of Turin (IT)
Publication Details
- Published
- 2025-06-30
- DOI
- https://doi.org/10.1109/icme59968.2025.11209228
- Citations
- 2
- Primary Topic
- 3D Shape Modeling and Analysis
- Type
- article
- Field-Weighted Citation Impact
- 2.67
Funders
- Grand Équipement National De Calcul Intensif