{"id":29194657,"url":"https://github.com/eidoslab/ddpm-pcc","last_synced_at":"2026-02-11T08:41:42.823Z","repository":{"id":301568233,"uuid":"958694326","full_name":"EIDOSLAB/DDPM-PCC","owner":"EIDOSLAB","description":null,"archived":false,"fork":false,"pushed_at":"2025-06-30T12:50:24.000Z","size":1073,"stargazers_count":2,"open_issues_count":0,"forks_count":0,"subscribers_count":4,"default_branch":"main","last_synced_at":"2025-09-09T06:58:15.867Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/EIDOSLAB.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2025-04-01T15:59:25.000Z","updated_at":"2025-09-06T01:38:41.000Z","dependencies_parsed_at":"2025-06-27T15:27:37.993Z","dependency_job_id":"6b4ca0fd-9964-45c8-b073-bba933b62e79","html_url":"https://github.com/EIDOSLAB/DDPM-PCC","commit_stats":null,"previous_names":["eidoslab/ddpm-pcc"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/EIDOSLAB/DDPM-PCC","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/EIDOSLAB%2FDDPM-PCC","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/EIDOSLAB%2FDDPM-PCC/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/EIDOSLAB%2FDDPM-PCC/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/EIDOSLAB%2FDDPM-PCC/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/EIDOSLAB","download_url":"https://codeload.github.com/EIDOSLAB/DDPM-PCC/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/EIDOSLAB%2FDDPM-PCC/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":29330174,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-02-11T06:13:03.264Z","status":"ssl_error","status_checked_at":"2026-02-11T06:12:55.843Z","response_time":97,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2025-07-02T04:06:07.880Z","updated_at":"2026-02-11T08:41:42.805Z","avatar_url":"https://github.com/EIDOSLAB.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Denoising Diffusion Probabilistic Model for Point Cloud Compression at Low Bit-Rates\n\nPytorch implementation of the paper \"**Denoising Diffusion Probabilistic Model for Point Cloud Compression at Low Bit-Rates**\", published at ICME 2025. This repository is based on [diffusion-point-cloud](https://github.com/luost26/diffusion-point-cloud).\n\n[ArXiv version](https://arxiv.org/pdf/2505.13316?)\n\n\n\n## Abstract\nEfficient compression of low-bit-rate point clouds is\ncritical for bandwidth-constrained applications. However, existing\ntechniques mainly focus on high-fidelity reconstruction, requiring\nmany bits for compression. This paper proposes a “Denoising\nDiffusion Probabilistic Model” (DDPM) architecture for point\ncloud compression (DDPM-PCC) at low bit-rates. A PointNet\nencoder produces the condition vector for the generation, which\nis then quantized via a learnable vector quantizer. This configuration allows to achieve a low bitrates while preserving\nquality. Experiments on ShapeNet and ModelNet40 show improved rate-distortion at low rates compared to standardized\nand state-of-the-art approaches\n\n\u003cdiv align=\"center\"\u003e\n\u003cimg src=\"imgs/arch.png\" alt=\"arch\" width=\"1200\"/\u003e\n\u003c/div\u003e\n\n\n## Usage\n\n### Environment\n- conda env create -f env.yml\n- conda activate point\n\n\n### Eval\nDownload our pretrained model in the DDPM-PCC directory from [here](https://drive.google.com/file/d/1c1J2G4x5Zr_nIeQT4Qh2ga1Jtz1H9nj2/view?usp=sharing).\n- Extract results_ddpm.zip\n```\nDDPM-PCC\n│   README.md\n│  \n└───results\n│   └───shapenet\n│       │   ...\n│   └───modelnet\n│       │   ...\n│   \n└───src\n    │   train.py\n    │   ...\n```\n\n- Run:\n```\nsh run_all_eval.sh\n```\n\n\n### Train\nDownload Shapenet and ModelNet datasets in DDPM-PCC directory from [here](https://drive.google.com/file/d/1eDYUmGA5dXTlLkuf7S25vQ4w17lS-5SS/view?usp=sharing).\n- Extract datasets.zip\n```\nDDPM-PCC\n│   README.md\n│  \n└───datasets\n│   └───modelnet40_ply_hdf5_2048/\n│       │   ...\n│   └───shapenet.hdf5\n│   \n└───src\n    │   train.py\n    │   ...\n```\n\n- Run:\n**ShapeNet**\n```\ncd src\n\npython train.py \\\n--dataset-path ../datasets/data/shapenet.hdf5 \\\n--dataset shapenet \\\n--latent-dim 256 \\\n--num-steps 200 \\\n--rotate 0 \\\n--save-dir ../results/shapenet/vq_diffusion/ae_all_pointnet_bs128 \\\n--encoder pointnet  \\\n--train-batch-size 128 \\\n--val-batch-size 1 \\\n--num-codecs 128 \\\n--dim-codecs 32 \\\n--vq-alpha 2.5\n```\n\n\n## Results\n\n- Quantitavie Results\n\u003cdiv align=\"center\"\u003e\n\u003cimg src=\"imgs/rd.png\" alt=\"rd\" width=1200\"/\u003e\n\u003c/div\u003e\n\n- Qualitative Results\n\u003cdiv align=\"center\"\u003e\n\u003cimg src=\"imgs/qualitative.png\" alt=\"rd\" width=1200\"/\u003e\n\u003c/div\u003e\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Feidoslab%2Fddpm-pcc","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Feidoslab%2Fddpm-pcc","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Feidoslab%2Fddpm-pcc/lists"}