{"id":18682519,"url":"https://github.com/postech-cvlab/point-transformer","last_synced_at":"2025-04-05T02:11:10.380Z","repository":{"id":37467643,"uuid":"327538089","full_name":"POSTECH-CVLab/point-transformer","owner":"POSTECH-CVLab","description":"This is an unofficial implementation of the Point Transformer paper.","archived":false,"fork":false,"pushed_at":"2022-04-19T15:41:29.000Z","size":3529,"stargazers_count":555,"open_issues_count":36,"forks_count":103,"subscribers_count":6,"default_branch":"master","last_synced_at":"2025-03-29T01:13:26.367Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/POSTECH-CVLab.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2021-01-07T07:37:31.000Z","updated_at":"2025-03-27T11:19:40.000Z","dependencies_parsed_at":"2022-07-12T14:50:46.447Z","dependency_job_id":null,"html_url":"https://github.com/POSTECH-CVLab/point-transformer","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/POSTECH-CVLab%2Fpoint-transformer","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/POSTECH-CVLab%2Fpoint-transformer/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/POSTECH-CVLab%2Fpoint-transformer/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/POSTECH-CVLab%2Fpoint-transformer/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/POSTECH-CVLab","download_url":"https://codeload.github.com/POSTECH-CVLab/point-transformer/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247276189,"owners_count":20912288,"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":[],"created_at":"2024-11-07T10:12:13.741Z","updated_at":"2025-04-05T02:11:10.358Z","avatar_url":"https://github.com/POSTECH-CVLab.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Point Transformer\nThis repository reproduces [Point Transformer](https://arxiv.org/abs/2012.09164). \\\nThe codebase is provided by the first author of [Point Transformer](https://arxiv.org/abs/2012.09164).\n\n## Notes\n- For shape classification and part segmentation, please use paconv-codebase branch. After some testing, we will merge it into the master branch.\n\n---\n## Dependencies\n- Ubuntu: 18.04 or higher\n- PyTorch: 1.9.0 \n- CUDA: 11.1 \n- Hardware: 4GPUs (TITAN RTX) to reproduce [Point Transformer](https://arxiv.org/abs/2012.09164) \n- To create conda environment, command as follows:\n\n  ```\n  bash env_setup.sh pt\n  ```\n\n## Dataset preparation\n- Download S3DIS [dataset](https://drive.google.com/uc?export=download\u0026id=1KUxWagmEWnvMhEb4FRwq2Mj0aa3U3xUf) and symlink the paths to them as follows:\n\n     ```\n     mkdir -p dataset\n     ln -s /path_to_s3dis_dataset dataset/s3dis\n     ```\n\n## Usage\n- Shape classification on ModelNet40\n  - For now, please use paconv-codebase branch.\n- Part segmentation on ShapeNetPart\n  - For now, please use paconv-codebase branch.\n- Semantic segmantation on S3DIS Area 5\n  - Train\n\n    - Specify the gpu used in config and then do training:\n\n      ```\n      sh tool/train.sh s3dis pointtransformer_repro\n      ```\n\n  - Test\n\n    - Afer training, you can test the checkpoint as follows:\n\n      ```\n      CUDA_VISIBLE_DEVICES=0 sh tool/test.sh s3dis pointtransformer_repro\n      ```\n  ---\n## Experimental Results\n\n- Semanctic Segmentation on S3DIS Area 5\n\n  |Model | mAcc | OA | mIoU |\n  |-------| ------| ----| -------|\n  |Paper| 76.5 | 90.8 | 70.4 |\n  |Hengshuang's code | 76.8 | 90.4 | 70.0 |\n---\n## References\n\nIf you use this code, please cite [Point Transformer](https://arxiv.org/abs/2012.09164):\n```\n@inproceedings{zhao2021point,\n  title={Point transformer},\n  author={Zhao, Hengshuang and Jiang, Li and Jia, Jiaya and Torr, Philip HS and Koltun, Vladlen},\n  booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision},\n  pages={16259--16268},\n  year={2021}\n}\n```\n\n## Acknowledgement\nThe code is from the first author of [Point Transformer](https://arxiv.org/abs/2012.09164).\nWe also refer [PAConv repository](https://github.com/CVMI-Lab/PAConv).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpostech-cvlab%2Fpoint-transformer","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fpostech-cvlab%2Fpoint-transformer","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpostech-cvlab%2Fpoint-transformer/lists"}