[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-72033":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":9,"language":10,"languages":9,"totalLinesOfCode":9,"stars":11,"forks":12,"watchers":13,"openIssues":14,"contributorsCount":15,"subscribersCount":15,"size":15,"stars1d":16,"stars7d":17,"stars30d":18,"stars90d":15,"forks30d":15,"starsTrendScore":19,"compositeScore":20,"rankGlobal":9,"rankLanguage":9,"license":21,"archived":22,"fork":22,"defaultBranch":23,"hasWiki":22,"hasPages":22,"topics":24,"createdAt":9,"pushedAt":9,"updatedAt":25,"readmeContent":26,"aiSummary":27,"trendingCount":15,"starSnapshotCount":15,"syncStatus":28,"lastSyncTime":29,"discoverSource":30},72033,"sam-3d-objects","facebookresearch\u002Fsam-3d-objects","facebookresearch","SAM 3D Objects",null,"Python",6892,814,65,93,0,95,161,280,285,39.73,"Other",false,"main",[],"2026-06-12 02:02:57","# SAM 3D\n\nSAM 3D Objects is one part of SAM 3D, a pair of models for object and human mesh reconstruction.  If you’re looking for SAM 3D Body, [click here](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fsam-3d-body).\n\n# SAM 3D Objects\n\n**SAM 3D Team**, [Xingyu Chen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=gjSHr6YAAAAJ&hl=en&oi=sra)\\*, [Fu-Jen Chu](https:\u002F\u002Ffujenchu.github.io\u002F)\\*, [Pierre Gleize](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=4imOcw4AAAAJ&hl=en&oi=ao)\\*, [Kevin J Liang](https:\u002F\u002Fkevinjliang.github.io\u002F)\\*, [Alexander Sax](https:\u002F\u002Falexsax.github.io\u002F)\\*, [Hao Tang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=XY6Nh9YAAAAJ&hl=en&oi=sra)\\*, [Weiyao Wang](https:\u002F\u002Fsites.google.com\u002Fview\u002Fweiyaowang\u002Fhome)\\*, [Michelle Guo](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=lyjjpNMAAAAJ&hl=en&oi=ao), [Thibaut Hardin](https:\u002F\u002Fgithub.com\u002FThibaut-H), [Xiang Li](https:\u002F\u002Fryanxli.github.io\u002F)⚬, [Aohan Lin](https:\u002F\u002Fgithub.com\u002Flinaohan), [Jia-Wei Liu](https:\u002F\u002Fjia-wei-liu.github.io\u002F), [Ziqi Ma](https:\u002F\u002Fziqi-ma.github.io\u002F)⚬, [Anushka Sagar](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fanushkasagar\u002F), [Bowen Song](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=QQKVkfcAAAAJ&hl=en&oi=sra)⚬, [Xiaodong Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?authuser=2&user=rMpcFYgAAAAJ), [Jianing Yang](https:\u002F\u002Fjedyang.com\u002F)⚬, [Bowen Zhang](http:\u002F\u002Fhome.ustc.edu.cn\u002F~zhangbowen\u002F)⚬, [Piotr Dollár](https:\u002F\u002Fpdollar.github.io\u002F)†, [Georgia Gkioxari](https:\u002F\u002Fgeorgiagkioxari.com\u002F)†, [Matt Feiszli](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=A-wA73gAAAAJ&hl=en&oi=ao)†§, [Jitendra Malik](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~malik\u002F)†§\n\n***Meta Superintelligence Labs***\n\n*Core contributor (Alphabetical, Equal Contribution), ⚬Intern, †Project leads, §Equal Contribution\n\n[[`Paper`](https:\u002F\u002Fai.meta.com\u002Fresearch\u002Fpublications\u002Fsam-3d-3dfy-anything-in-images\u002F)] [[`Code`](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fsam-3d-objects)] [[`Website`](https:\u002F\u002Fai.meta.com\u002Fsam3d\u002F)] [[`Demo`](https:\u002F\u002Fwww.aidemos.meta.com\u002Fsegment-anything\u002Feditor\u002Fconvert-image-to-3d)] [[`Blog`](https:\u002F\u002Fai.meta.com\u002Fblog\u002Fsam-3d\u002F)] [[`BibTeX`](#citing-sam-3d-objects)] [[`Roboflow`](https:\u002F\u002Fblog.roboflow.com\u002Fsam-3d\u002F)]\n\n**SAM 3D Objects** is a foundation model that reconstructs full 3D shape geometry, texture, and layout from a single image, excelling in real-world scenarios with occlusion and clutter by using progressive training and a data engine with human feedback. It outperforms prior 3D generation models in human preference tests on real-world objects and scenes. We released code, weights, online demo, and a new challenging benchmark.\n\n\n\u003Cp align=\"center\">\u003Cimg src=\"doc\u002Fintro.png\"\u002F>\u003C\u002Fp>\n\n-----\n\n\u003Cp align=\"center\">\u003Cimg src=\"doc\u002Farch.png\"\u002F>\u003C\u002Fp>\n\n## Latest updates\n\n**11\u002F19\u002F2025** - Checkpoints Launched, Web Demo and Paper are out.\n\n## Installation\n\nFollow the [setup](doc\u002Fsetup.md) steps before running the following.\n\n## Single or Multi-Object 3D Generation\n\nSAM 3D Objects can convert masked objects in an image, into 3D models with pose, shape, texture, and layout. SAM 3D is designed to be robust in challenging natural images, handling small objects and occlusions, unusual poses, and difficult situations encountered in uncurated natural scenes like this kidsroom:\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"notebook\u002Fimages\u002Fshutterstock_stylish_kidsroom_1640806567\u002Fimage.png\" width=\"55%\"\u002F>\n  \u003Cimg src=\"doc\u002Fkidsroom_transparent.gif\" width=\"40%\"\u002F>\n\u003C\u002Fp>\n\nFor a quick start, run `python demo.py` or use the the following lines of code:\n\n```python\nimport sys\n\n# import inference code\nsys.path.append(\"notebook\")\nfrom inference import Inference, load_image, load_single_mask\n\n# load model\ntag = \"hf\"\nconfig_path = f\"checkpoints\u002F{tag}\u002Fpipeline.yaml\"\ninference = Inference(config_path, compile=False)\n\n# load image and mask\nimage = load_image(\"notebook\u002Fimages\u002Fshutterstock_stylish_kidsroom_1640806567\u002Fimage.png\")\nmask = load_single_mask(\"notebook\u002Fimages\u002Fshutterstock_stylish_kidsroom_1640806567\", index=14)\n\n# run model\noutput = inference(image, mask, seed=42)\n\n# export gaussian splat\noutput[\"gs\"].save_ply(f\"splat.ply\")\n```\n\nFor  more details and multi-object reconstruction, please take a look at out two jupyter notebooks:\n* [single object](notebook\u002Fdemo_single_object.ipynb)\n* [multi object](notebook\u002Fdemo_multi_object.ipynb)\n\n\n## SAM 3D Body\n\n[SAM 3D Body (3DB)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fsam-3d-body) is a robust promptable foundation model for single-image 3D human mesh recovery (HMR).\n\nAs a way to combine the strengths of both **SAM 3D Objects** and **SAM 3D Body**, we provide an example notebook that demonstrates how to combine the results of both models such that they are aligned in the same frame of reference. Check it out [here](notebook\u002Fdemo_3db_mesh_alignment.ipynb).\n\n## License\n\nThe SAM 3D Objects model checkpoints and code are licensed under [SAM License](.\u002FLICENSE).\n\n## Contributing\n\nSee [contributing](CONTRIBUTING.md) and the [code of conduct](CODE_OF_CONDUCT.md).\n\n## Contributors\n\nThe SAM 3D Objects project was made possible with the help of many contributors.\n\nRobbie Adkins,\nParis Baptiste,\nKaren Bergan,\nKai Brown,\nMichelle Chan,\nIda Cheng,\nKhadijat Durojaiye,\nPatrick Edwards,\nDaniella Factor,\nFacundo Figueroa,\nRene  de la Fuente,\nEva Galper,\nCem Gokmen,\nAlex He,\nEnmanuel Hernandez,\nDex Honsa,\nLeonna Jones,\nArpit Kalla,\nKris Kitani,\nHelen Klein,\nKei Koyama,\nRobert Kuo,\nVivian Lee,\nAlex Lende,\nJonny Li,\nKehan Lyu,\nFaye Ma,\nMallika Malhotra,\nSasha Mitts,\nWilliam Ngan,\nGeorge Orlin,\nPeter Park,\nDon Pinkus,\nRoman Radle,\nNikhila Ravi,\nAzita Shokrpour,\nJasmine Shone,\nZayida Suber,\nPhillip Thomas,\nTatum Turner,\nJoseph Walker,\nMeng Wang,\nClaudette Ward,\nAndrew Westbury,\nLea Wilken,\nNan Yang,\nYael Yungster\n\n\n## Citing SAM 3D Objects\n\nIf you use SAM 3D Objects in your research, please use the following BibTeX entry.\n\n```\n@article{sam3dteam2025sam3d3dfyimages,\n      title={SAM 3D: 3Dfy Anything in Images}, \n      author={SAM 3D Team and Xingyu Chen and Fu-Jen Chu and Pierre Gleize and Kevin J Liang and Alexander Sax and Hao Tang and Weiyao Wang and Michelle Guo and Thibaut Hardin and Xiang Li and Aohan Lin and Jiawei Liu and Ziqi Ma and Anushka Sagar and Bowen Song and Xiaodong Wang and Jianing Yang and Bowen Zhang and Piotr Dollár and Georgia Gkioxari and Matt Feiszli and Jitendra Malik},\n      year={2025},\n      eprint={2511.16624},\n      archivePrefix={arXiv},\n      primaryClass={cs.CV},\n      url={https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.16624}, \n}\n```\n","SAM 3D Objects 是一个从单张图片中重建完整3D形状几何、纹理和布局的基础模型。该项目利用渐进式训练和包含人类反馈的数据引擎，在处理遮挡和杂乱的真实场景时表现出色，超越了之前的3D生成模型在真实物体和场景上的人类偏好测试中的表现。它适合用于需要将2D图像转换为高质量3D模型的应用场景，如虚拟现实、增强现实以及游戏开发等。项目提供了代码、权重、在线演示及新的挑战基准，支持单个或多个对象的3D生成。",2,"2026-06-11 03:40:01","high_star"]