{"id":20857799,"url":"https://github.com/zju3dv/manhattan_sdf","last_synced_at":"2025-04-05T14:08:43.401Z","repository":{"id":38233393,"uuid":"489086984","full_name":"zju3dv/manhattan_sdf","owner":"zju3dv","description":"Code for \"Neural 3D Scene Reconstruction with the Manhattan-world Assumption\" CVPR 2022 Oral","archived":false,"fork":false,"pushed_at":"2023-06-06T12:48:58.000Z","size":747,"stargazers_count":484,"open_issues_count":1,"forks_count":34,"subscribers_count":23,"default_branch":"main","last_synced_at":"2024-05-18T19:43:00.476Z","etag":null,"topics":["3d-reconstruction","3d-vision","computer-vision","cvpr2022"],"latest_commit_sha":null,"homepage":"https://zju3dv.github.io/manhattan_sdf/","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/zju3dv.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}},"created_at":"2022-05-05T18:36:01.000Z","updated_at":"2024-05-13T02:05:52.000Z","dependencies_parsed_at":"2024-04-21T09:49:39.513Z","dependency_job_id":"5033ce3d-2061-43f3-82f5-fd4771a4b772","html_url":"https://github.com/zju3dv/manhattan_sdf","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/zju3dv%2Fmanhattan_sdf","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zju3dv%2Fmanhattan_sdf/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zju3dv%2Fmanhattan_sdf/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zju3dv%2Fmanhattan_sdf/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/zju3dv","download_url":"https://codeload.github.com/zju3dv/manhattan_sdf/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247345854,"owners_count":20924102,"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-reconstruction","3d-vision","computer-vision","cvpr2022"],"created_at":"2024-11-18T04:43:12.859Z","updated_at":"2025-04-05T14:08:43.373Z","avatar_url":"https://github.com/zju3dv.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"**News**\n\n* `06/03/2022` We provide the instruction to run on custom data [here](./docs/CUSTOM.md).\n* `05/10/2022` To make the comparison on ScanNet easier, we provide all quantitative and qualitative results of baselines [here](./docs/RESULTS.md#baseline-results), including COLMAP, COLMAP*, ACMP, NeRF, UNISURF, NeuS, and VolSDF.\n* `05/10/2022` To make the following works easier to compare with our model, we provide our quantitative and qualitative results, as well as the trained models on ScanNet [here](./docs/RESULTS.md#our-results).\n* `05/10/2022` We upload our processed ScanNet scene data to [Google Drive](https://drive.google.com/drive/folders/1LyeLuJzTH7-JWUG886gpN1WEzxGDRLm3?usp=sharing).\n\n# Neural 3D Scene Reconstruction with the Manhattan-world Assumption\n### [Project Page](https://zju3dv.github.io/manhattan_sdf) | [Video](https://www.youtube.com/watch?v=U4zmSuh31g0) | [Paper](https://arxiv.org/abs/2205.02836)\n\u003cbr/\u003e\n\n![introduction](./assets/introduction.png)\n\n\u003e [Neural 3D Scene Reconstruction with the Manhattan-world Assumption](https://arxiv.org/abs/2205.02836)  \n\u003e [Haoyu Guo](https://github.com/ghy0324)\u003csup\u003e\\*\u003c/sup\u003e, [Sida Peng](https://pengsida.net)\u003csup\u003e\\*\u003c/sup\u003e, [Haotong Lin](https://github.com/haotongl), [Qianqian Wang](http://www.cs.cornell.edu/~qqw/), [Guofeng Zhang](http://www.cad.zju.edu.cn/home/gfzhang/), [Hujun Bao](http://www.cad.zju.edu.cn/home/bao/), [Xiaowei Zhou](https://xzhou.me)  \n\u003e CVPR 2022 (Oral Presentation)\n\u003cbr/\u003e\n\n## Setup\n\n### Installation\n```shell\nconda env create -f environment.yml\nconda activate manhattan\n```\n\n### Data preparation\n\nDownload ScanNet scene data evaluated in the paper from [Google Drive](https://drive.google.com/drive/folders/1LyeLuJzTH7-JWUG886gpN1WEzxGDRLm3?usp=sharing) and extract them into `data/`. Make sure that the path is consistent with [config file](./configs/scannet/0050.yaml#L38).\n\nWe provide the instruction to run on custom data [here](./docs/CUSTOM.md).\n\n## Usage\n\n### Training\n\n```shell\npython train_net.py --cfg_file configs/scannet/0050.yaml gpus 0, exp_name scannet_0050\n```\n\n### Mesh extraction\n\n```shell\npython run.py --type mesh_extract --output_mesh result.obj --cfg_file configs/scannet/0050.yaml gpus 0, exp_name scannet_0050\n```\n\n### Evaluation\n\n```shell\npython run.py --type evaluate --cfg_file configs/scannet/0050.yaml gpus 0, exp_name scannet_0050\n```\n\n\n## Citation\n\nIf you find this code useful for your research, please use the following BibTeX entry.\n\n```bibtex\n@inproceedings{guo2022manhattan,\n  title={Neural 3D Scene Reconstruction with the Manhattan-world Assumption},\n  author={Guo, Haoyu and Peng, Sida and Lin, Haotong and Wang, Qianqian and Zhang, Guofeng and Bao, Hujun and Zhou, Xiaowei},\n  booktitle={CVPR},\n  year={2022}\n}\n```\n\n## Acknowledgement\n\n- Thanks to Lior Yariv for her excellent work [VolSDF](https://lioryariv.github.io/volsdf/).\n- Thanks to Jianfei Guo for his implementation of VolSDF [neurecon](https://github.com/ventusff/neurecon).\n- Thanks to Johannes Schönberger for his excellent work [COLMAP](https://github.com/colmap/colmap).\n- Thanks to Shaohui Liu for his customized implementation of [COLMAP](https://github.com/B1ueber2y/colmap/tree/c84269d693246d8294307cc32f851813f18b6a2d) as a submodule of [NerfingMVS](https://github.com/weiyithu/NerfingMVS).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fzju3dv%2Fmanhattan_sdf","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fzju3dv%2Fmanhattan_sdf","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fzju3dv%2Fmanhattan_sdf/lists"}