{"id":18710483,"url":"https://github.com/vision-cair/3dcompat-v2","last_synced_at":"2025-04-15T18:32:57.127Z","repository":{"id":154184869,"uuid":"627716663","full_name":"Vision-CAIR/3DCoMPaT-v2","owner":"Vision-CAIR","description":"3DCoMPaT++: An improved large-scale 3D vision dataset for compositional recognition","archived":false,"fork":false,"pushed_at":"2024-07-09T17:02:27.000Z","size":139856,"stargazers_count":82,"open_issues_count":1,"forks_count":6,"subscribers_count":7,"default_branch":"main","last_synced_at":"2025-03-28T23:34:39.087Z","etag":null,"topics":["3d","compositional-learning","computer-vision","deep-learning","multimodal-deep-learning"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"bsd-3-clause","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Vision-CAIR.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":"2023-04-14T03:37:52.000Z","updated_at":"2025-03-21T09:45:08.000Z","dependencies_parsed_at":null,"dependency_job_id":"40df4b23-bf6f-4174-9dcf-14df888011e5","html_url":"https://github.com/Vision-CAIR/3DCoMPaT-v2","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Vision-CAIR%2F3DCoMPaT-v2","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Vision-CAIR%2F3DCoMPaT-v2/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Vision-CAIR%2F3DCoMPaT-v2/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Vision-CAIR%2F3DCoMPaT-v2/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Vision-CAIR","download_url":"https://codeload.github.com/Vision-CAIR/3DCoMPaT-v2/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":249129253,"owners_count":21217314,"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","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","compositional-learning","computer-vision","deep-learning","multimodal-deep-learning"],"created_at":"2024-11-07T12:34:21.618Z","updated_at":"2025-04-15T18:32:52.117Z","avatar_url":"https://github.com/Vision-CAIR.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cdiv align=\"center\"\u003e\n\u003cp align=\"center\"\u003e\n     \u003cimg src=\"img/logo.png\" width=500px/\u003e\n\u003c/p\u003e \n\u003ch1 align=\"center\"\u003e\n\u003c/h1\u003e\n\u003ch1 align=\"center\"\u003e\n    3DCoMPaT++: An improved Large-scale 3D Vision Dataset for Compositional Recognition\n\u003c/h1\u003e\n\n[![PAper](https://img.shields.io/badge/Paper-red?logo=arxiv\u0026logoWidth=15)](https://arxiv.org/abs/2310.18511)\n[![Jupyter Quickstart](https://img.shields.io/badge/Quickstart-orange?logo=google-colab\u0026logoWidth=15)](https://colab.research.google.com/drive/1OpgYL_cxekAqZF8B8zuQZkPQxUIxzV0K?usp=sharing)\n[![Documentation](https://img.shields.io/badge/📚%20Documentation-blue?logoColor=white\u0026logoWidth=20)](https://3dcompat-dataset.org/doc/)\n[![Download](https://img.shields.io/badge/📦%20Download-grey?logoColor=white\u0026logoWidth=20)](https://3dcompat-dataset.org/doc/dl-dataset.html)\n[![Website](https://img.shields.io/badge/🌐%20Website-green?logoColor=white\u0026logoWidth=20)](https://3dcompat-dataset.org/)\n[![Workshop](https://img.shields.io/badge/🔨%20Workshop-purple?logoColor=white\u0026logoWidth=20)](https://3dcompat-dataset.org/workshop/)\n[![Challenge](https://img.shields.io/badge/🏆%20Challenge-critical?logoColor=white\u0026logoWidth=20)](https://eval.ai/web/challenges/challenge-page/2031)\n\n\u003c/div\u003e\n\n## 📰 News\n\n- **19/08/2023**: As our CVPR23 challenge has finished (congratulations to [Cattalyya Nuengsikapian](https://3dcompat-dataset.org/workshop/#main-section)!), our test set has now been made public. Dataloaders have been updated in consequence: using the \"`EvalLoader`\" classes is not necessary anymore 😊\n\n- **18/06/2023**: The 3DCoMPaT++ CVPR23 challenge has been concluded. We would like to congratulate [Cattalyya Nuengsikapian](https://3dcompat-dataset.org/workshop/#main-section), winner of both **coarse** and **fine-grained** tracks for her excellent performance in our challenge 🎉\n\n## Summary\n\n- [Introduction](#📚-introduction)\n- [Getting started](#🚀-getting-started)\n- [Baselines](#📊-baselines)\n- [Challenge](#🏆-challenge)\n- [Acknowledgments](#🙏-acknowledgments)\n- [Citation](#citation)\n\n\u003cbr\u003e\n\n![3DCoMPaT models view](img/header_gif.gif)\n\n\u003cbr\u003e\n\n## 📚 Introduction\n\n3DCoMPaT++ is a multimodal 2D/3D dataset of 16 million rendered views of more than 10 million stylized 3D shapes carefully annotated at **part-instance** level, alongside matching **RGB pointclouds**, **3D textured meshes**, **depth maps** and **segmentation masks**. This work builds upon [3DCoMPaT](https://3dcompat-dataset.org/), the first version of this dataset.\n\n**We plan to further extend the dataset: stay tuned!** 🔥\n\n\u003cbr\u003e\n\n## 🔍 Browser\n\nTo explore our dataset, please check out our integrated web browser:\n\n\u003ca href=\"https://3dcompat-dataset.org/browser\"\u003e\n    \u003cp align=\"center\"\u003e\n    \u003cimg src=\"img/browser_sticker.png\"\n        alt=\"3DCoMPaT Browser\"\n        style=\"width:600px;\" /\u003e\n    \u003c/p\u003e\n\u003c/a\u003e\n\nFor more information about the shape browser, please check out [our dedicated Wiki page](https://3dcompat-dataset.org/doc/browser.html).\n\n\u003cbr\u003e\n\n## 🚀 Getting started\n\nTo get started straight away, here is a Jupyter notebook (no downloads required, just **run and play**!):\n\n[![Jupyter Quickstart](https://img.shields.io/badge/Quickstart-orange?logo=google-colab\u0026logoWidth=15)](https://colab.research.google.com/drive/1OpgYL_cxekAqZF8B8zuQZkPQxUIxzV0K?usp=sharing)\n\nFor a deeper dive into our dataset, please check our online documentation:\n\n[![Documentation](https://img.shields.io/badge/📚%20Documentation-blue?logoColor=white)](https://3dcompat-dataset.org/doc/)\n\n\u003cbr\u003e\n\n## 📊 Baselines\n\nWe provide baseline models for 2D and 3D tasks, following the structure below:\n\n- **2D Experiments**\n  - [2D Shape Classifier](./models/2D/shape_classifier/): ResNet50\n  - [2D Part and Material Segmentation](./models/2D/segmentation/): SegFormer\n- **3D Experiments**\n  - [3D Shape classification](./models/3D/): DGCNN - PCT - PointNet++ - PointStack - Curvenet - PointNext - PointMLP\n  - [3D Part segmentation](./models/3D/): PCT - PointNet++ - PointStack - Curvenet - PointNeXT\n\n\u003cbr\u003e\n\n## 🏆 Challenge\n\nAs a part of the [C3DV CVPR 2023 workshop](https://3dcompat-dataset.org/workshop/), we are organizing a modelling challenge based on 3DCoMPaT++.\nTo learn more about the challenge, check out this link:\n\n[![Challenge](https://img.shields.io/badge/🏆%20Challenge-critical?logoColor=white\u0026logoWidth=20)](https://eval.ai/web/challenges/challenge-page/2031)\n\n\u003cbr\u003e\n\n## 🙏 Acknowledgments\n\n⚙️ For computer time, this research used the resources of the Supercomputing Laboratory at [King Abdullah University of Science \u0026 Technology (KAUST)](https://www.kaust.edu.sa/).\nWe extend our sincere gratitude to the [KAUST HPC Team](www.hpc.kaust.edu.sa) for their invaluable assistance and support during the course of this research project. Their expertise and dedication continues to play a crucial role in the success of our work.\n\n💾 We also thank the [Amazon Open Data](https://aws.amazon.com/opendata) program for providing us with free storage of our large-scale data on their servers. Their generosity and commitment to making research data widely accessible have greatly facilitated our research efforts.\n\n\u003c/br\u003e\n\n## Citation\n\nIf you use our dataset, please cite the two following references:\n\n```bibtex\n@article{slim2023_3dcompatplus,\n    title={3DCoMPaT++: An improved Large-scale 3D Vision Dataset\n    for Compositional Recognition},\n    author={Habib Slim, Xiang Li, Yuchen Li,\n    Mahmoud Ahmed, Mohamed Ayman, Ujjwal Upadhyay\n    Ahmed Abdelreheem, Arpit Prajapati,\n    Suhail Pothigara, Peter Wonka, Mohamed Elhoseiny},\n    year={2023}\n}\n```\n\n```bibtex\n@article{li2022_3dcompat,\n    title={3D CoMPaT: Composition of Materials on Parts of 3D Things},\n    author={Yuchen Li, Ujjwal Upadhyay, Habib Slim,\n    Ahmed Abdelreheem, Arpit Prajapati,\n    Suhail Pothigara, Peter Wonka, Mohamed Elhoseiny},\n    journal = {ECCV},\n    year={2022}\n}\n```\n\n\u003c/br\u003e\n\nThis repository is owned and maintained by \u003ca href=\"https://habibslim.github.io/\"\u003eHabib Slim\u003c/a\u003e, \u003ca href=\"https://xiangli.ac.cn/\"\u003eXiang Li\u003c/a\u003e, \u003ca href=\"mahmoudalsayed@aucegypt.edu\"\u003eMahmoud Ahmed\u003c/a\u003e and \u003ca href=\"https://personal-website-mohamedayman15069.vercel.app/\"\u003eMohamed Ayman\u003c/a\u003e, from the \u003ca href=\"https://cemse.kaust.edu.sa/vision-cair\"\u003eVision-CAIR\u003c/a\u003e group.\n\n## References\n\n1. _[Li et al., 2022]_ - 3DCoMPaT: Composition of Materials on Parts of 3D Things.\n2. _[Xie et al., 2021]_ - SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers.\n3. _[He et al., 2015]_ - Deep Residual Learning for Image Recognition.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fvision-cair%2F3dcompat-v2","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fvision-cair%2F3dcompat-v2","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fvision-cair%2F3dcompat-v2/lists"}