{"id":13444079,"url":"https://github.com/pqhieu/jsis3d","last_synced_at":"2025-08-30T03:26:59.455Z","repository":{"id":123086773,"uuid":"179839998","full_name":"pqhieu/jsis3d","owner":"pqhieu","description":"[CVPR'19] JSIS3D: Joint Semantic-Instance Segmentation of 3D Point Clouds","archived":false,"fork":false,"pushed_at":"2020-07-08T07:49:31.000Z","size":1558,"stargazers_count":177,"open_issues_count":2,"forks_count":35,"subscribers_count":6,"default_branch":"master","last_synced_at":"2025-03-20T17:41:51.137Z","etag":null,"topics":["cvpr","deep-learning","instance-segmentation","point-cloud","pytorch"],"latest_commit_sha":null,"homepage":"https://pqhieu.com/research/cvpr19/","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/pqhieu.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}},"created_at":"2019-04-06T13:58:48.000Z","updated_at":"2025-03-14T08:14:19.000Z","dependencies_parsed_at":"2023-06-15T14:30:58.494Z","dependency_job_id":null,"html_url":"https://github.com/pqhieu/jsis3d","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/pqhieu/jsis3d","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pqhieu%2Fjsis3d","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pqhieu%2Fjsis3d/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pqhieu%2Fjsis3d/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pqhieu%2Fjsis3d/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/pqhieu","download_url":"https://codeload.github.com/pqhieu/jsis3d/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pqhieu%2Fjsis3d/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":272799979,"owners_count":24995104,"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":"2025-08-30T02:00:09.474Z","response_time":77,"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":["cvpr","deep-learning","instance-segmentation","point-cloud","pytorch"],"created_at":"2024-07-31T03:02:18.440Z","updated_at":"2025-08-30T03:26:59.428Z","avatar_url":"https://github.com/pqhieu.png","language":"Python","funding_links":[],"categories":["Python"],"sub_categories":[],"readme":"# JSIS3D\n\nThis is the official Pytorch implementation of the following publication.\n\n\u003e **JSIS3D: Joint Semantic-Instance Segmentation of 3D Point Clouds with**\u003cbr/\u003e\n\u003e **Multi-Task Pointwise Networks and Multi-Value Conditional Random Fields**\u003cbr/\u003e\n\u003e Quang-Hieu Pham, Duc Thanh Nguyen, Binh-Son Hua, Gemma Roig, Sai-Kit\n\u003e Yeung\u003cbr/\u003e *Conference on Computer Vision and Pattern Recognition (CVPR),\n\u003e 2019* (**Oral**)\u003cbr/\u003e\n\u003e [Paper](https://arxiv.org/abs/1904.00699) |\n\u003e [Homepage](https://pqhieu.github.io/research/cvpr19/)\n\n### Citation\nIf you find our work useful for your research, please consider citing:\n\n    @inproceedings{pham-jsis3d-cvpr19,\n      title = {{JSIS3D}: Joint semantic-instance segmentation of 3d point clouds with multi-task pointwise networks and multi-value conditional random fields},\n      author = {Pham, Quang-Hieu and Nguyen, Duc Thanh and Hua, Binh-Son and Roig, Gemma and Yeung, Sai-Kit},\n      booktitle = {Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},\n      year = {2019}\n    }\n\n## Usage\n\n### Prerequisites\nThis code is tested in Manjaro Linux with CUDA 10.0 and Pytorch 1.0.\n\n- Python 3.5+\n- Pytorch 0.4.0+\n\n### Installation\nTo use MV-CRF (optional), you first need to compile the code:\n\n    cd external/densecrf\n    mkdir build\n    cd build\n    cmake -D CMAKE_BUILD_TYPE=Release ..\n    make\n    cd ../../.. # You should be at the root folder here\n    make\n\n### Dataset\nWe have preprocessed the S3DIS dataset ([2.5GB](https://drive.google.com/open?id=1s1cFfb8cInM-SNHQoTGxN9BIyNpNQK6x))\nin HDF5 format. After downloading the files, put them into the corresponding\n`data/s3dis/h5` folder.\n\n### Training \u0026 Evaluation\nTo train a model on S3DIS dataset:\n\n    python train.py --config configs/s3dis.json --logdir logs/s3dis\n\nLog files and network parameters will be saved to the `logs/s3dis` folder.\n\nAfter training, we can use the model to predict semantic-instance segmentation\nlabels as follows:\n\n    python pred.py --logdir logs/s3dis --mvcrf\n\nTo evaluate the results, run the following command:\n\n    python eval.py --logdir logs/s3dis\n\nFor more details, you can use the `--help` option for every scripts.\n\n\u003e **Note**: The results on S3DIS in our paper are tested on Area 6 instead of Area 5.\n\u003e To reproduce the results, please change the split in `train.txt` and `test.txt` accordingly.\n\u003e Here I chose to keep the test set on Area 5 to make it easier to compare with other methods.\n\n### Prepare your own dataset\nCheck out the `scripts` folder to see how we prepare the dataset for training.\n\n## License\nOur code is released under MIT license (see LICENSE for more details).\n\n**Contact**: Quang-Hieu Pham (pqhieu1192@gmail.com)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpqhieu%2Fjsis3d","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fpqhieu%2Fjsis3d","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpqhieu%2Fjsis3d/lists"}