{"id":24785482,"url":"https://github.com/qizekun/VPP","last_synced_at":"2025-10-12T09:30:53.142Z","repository":{"id":196858809,"uuid":"673178814","full_name":"qizekun/VPP","owner":"qizekun","description":"[NeurIPS 2023] VPP: Efficient Conditional 3D Generation via Voxel-Point Progressive Representation","archived":false,"fork":false,"pushed_at":"2024-06-27T15:07:59.000Z","size":3886,"stargazers_count":32,"open_issues_count":2,"forks_count":1,"subscribers_count":6,"default_branch":"main","last_synced_at":"2025-01-29T14:01:53.918Z","etag":null,"topics":["3d","3d-generation","3d-point-clouds","conditional-generation","representation-learning"],"latest_commit_sha":null,"homepage":"https://arxiv.org/abs/2307.16605","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/qizekun.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}},"created_at":"2023-08-01T03:48:57.000Z","updated_at":"2025-01-23T03:56:40.000Z","dependencies_parsed_at":"2025-01-29T14:01:59.890Z","dependency_job_id":null,"html_url":"https://github.com/qizekun/VPP","commit_stats":null,"previous_names":["qizekun/vpp"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/qizekun/VPP","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/qizekun%2FVPP","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/qizekun%2FVPP/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/qizekun%2FVPP/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/qizekun%2FVPP/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/qizekun","download_url":"https://codeload.github.com/qizekun/VPP/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/qizekun%2FVPP/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":279010939,"owners_count":26084837,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-10-12T02:00:06.719Z","response_time":53,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"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":["3d","3d-generation","3d-point-clouds","conditional-generation","representation-learning"],"created_at":"2025-01-29T14:01:48.566Z","updated_at":"2025-10-12T09:30:52.232Z","avatar_url":"https://github.com/qizekun.png","language":"Python","funding_links":[],"categories":["Python"],"sub_categories":[],"readme":"# VPP⚡\n\n\u003e [**VPP: Efficient Conditional 3D Generation via Voxel-Point Progressive Representation**](https://arxiv.org/abs/2307.16605), **NeurIPS 2023** \u003cbr\u003e\n\u003e [Zekun Qi](https://scholar.google.com/citations?user=ap8yc3oAAAAJ), [Muzhou Yu](https://github.com/muzhou-yu), [Runpei Dong](https://runpeidong.com/) and [Kaisheng Ma](http://group.iiis.tsinghua.edu.cn/~maks/leader.html) \u003cbr\u003e\n\n[arXiv Paper](https://arxiv.org/abs/2307.16605)\n\n## News\n\n- 🎆 Sep, 2023: [**VPP**](https://arxiv.org/abs/2307.16605) is accepted to NeurIPS 2023.\n- 💥 Aug, 2023: Check out our previous work [**ACT**](https://arxiv.org/abs/2212.08320) and [**ReCon**](https://arxiv.org/abs/2302.02318) about 3D represent learning, which have been accepted by ICLR \u0026 ICML 2023.\n\n----\n\n\nThis repository contains the code release of VPP⚡: Efficient Conditional 3D Generation via Voxel-Point Progressive Representation.\n\n\u003cdiv  align=\"center\"\u003e    \n \u003cimg src=\"./figure/generation.png\" width = \"1100\"  align=center /\u003e\n\u003c/div\u003e\n\n## 1. Requirements\nPyTorch \u003e= 1.7.0;\npython \u003e= 3.7;\nCUDA \u003e= 9.0;\nGCC \u003e= 4.9;\ntorchvision;\n\n```\n# Quick Start\nconda create -n vpp python=3.8 -y\nconda activate vpp\n\nconda install pytorch==1.10.0 torchvision==0.11.0 cudatoolkit=11.3 -c pytorch -c nvidia\n# pip install torch==1.10.0+cu113 torchvision==0.11.0+cu113 torchaudio==0.10.0+cu113 -f https://download.pytorch.org/whl/torch_stable.html\n```\n\n```\n# Install basic required packages\npip install -r requirements.txt\n# Chamfer Distance\ncd ./extensions/chamfer_dist \u0026\u0026 python setup.py install --user\n\n# install sap to recontruct meshes from point clouds\ncd sap\ncd pointnet2_ops_lib \u0026\u0026 pip install -e .\nwget https://github.com/facebookresearch/pytorch3d/archive/refs/tags/v0.6.1.zip\nunzip v0.6.1.zip\ncd pytorch3d-0.6.1/ \u0026\u0026 pip install -e .\n```\n\n## 2. Training\n\n### 2.1 Data Preparation\nSee [DATASET.md](./DATASET.md) for details.\n\n### 2.2 Training 3D VQGAN\n```\nsh scripts/train_vqgan.sh \u003cgpu_id\u003e\n```\n\n### 2.3 Training Voxel Semantic Generator\n```\nsh scripts/train_voxel_generator.sh \u003cgpu_id\u003e\n```\n\n### 2.4 Training Grid Smoother\n```\nsh scripts/train_grid_smoother.sh \u003cgpu_id\u003e\n```\n\n### 2.5 Training Point Upsampler\n```\nsh scripts/train_point_upsampler.sh \u003cgpu_id\u003e\n```\n\n## 3. Conditional Point Clouds Generation\n\ntext prompt:\n```\nsh scripts/inference/text_prompt.sh \u003cgpu_id\u003e \"a round table.\"\n```\n\nimage prompt:\n```\nsh scripts/inference/image_prompt.sh \u003cgpu_id\u003e \u003cimg_path\u003e\n```\n\n\n## 5. Reconstruction Meshes from Point Clouds\nDownload the [pretrained sap model](https://drive.google.com/file/d/1Ui44qxMueL21REtoeuCBwrdJPmpxegmZ/view?usp=drive_link).\n```\ncd sap\nexport CUDA_VISIBLE_DEVICES=0 \u0026\u0026 python mesh_reconstruction.py --config ../sap.json --ckpt ../sap.pkl --dataset_path ../points.npz --save_dir output/\n```\nThe [shape as points](https://arxiv.org/abs/2106.03452) reconstruction pipeline originates from [SLIDE](https://github.com/SLIDE-3D/SLIDE), which has been trained on multi-category ShapeNet datasets with artificial noise.\n\n\n## 6. Visualization\nWe use [PointVisualizaiton](https://github.com/qizekun/PointVisualizaiton) repo to render beautiful point cloud image, including specified color rendering and attention distribution rendering.\n\n## Contact\n\nIf you have any questions related to the code or the paper, feel free to email Zekun (`qizekun@gmail.com`). \n\n## License\n\nVPP is released under MIT License. See the [LICENSE](./LICENSE) file for more details.\n\n## Acknowledgements\n\nThis codebase is built upon [Point-MAE](https://github.com/Pang-Yatian/Point-MAE), [CLIP](https://github.com/openai/CLIP) and [SLIDE](https://github.com/SLIDE-3D/SLIDE)\n\n## Citation\n\nIf you find our work useful in your research, please consider citing:\n\n```latex\n@inproceedings{vpp2023,\ntitle={{VPP}: Efficient Universal 3D Generation via Voxel-Point Progressive Representation},\nauthor={Qi, Zekun and Yu, Muzhou and Dong, Runpei and Ma, Kaisheng},\nbooktitle={Thirty-seventh Conference on Neural Information Processing Systems},\nyear={2023},\nurl={https://openreview.net/forum?id=etd0ebzGOG}\n}\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fqizekun%2FVPP","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fqizekun%2FVPP","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fqizekun%2FVPP/lists"}