{"id":13439312,"url":"https://github.com/XuyangBai/KPConv.pytorch","last_synced_at":"2025-03-20T07:33:09.360Z","repository":{"id":117709475,"uuid":"228848276","full_name":"XuyangBai/KPConv.pytorch","owner":"XuyangBai","description":"PyTorch reimplementation for \"KPConv: Flexible and Deformable Convolution for Point Clouds\" https://arxiv.org/abs/1904.08889","archived":false,"fork":false,"pushed_at":"2020-07-09T01:26:17.000Z","size":113,"stargazers_count":66,"open_issues_count":2,"forks_count":9,"subscribers_count":6,"default_branch":"master","last_synced_at":"2024-10-28T01:12:09.130Z","etag":null,"topics":["3d-point-clouds","geometric-deep-learning","kpconv","pytorch"],"latest_commit_sha":null,"homepage":"","language":"C++","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/XuyangBai.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,"governance":null,"roadmap":null,"authors":null}},"created_at":"2019-12-18T13:40:57.000Z","updated_at":"2024-09-16T06:19:41.000Z","dependencies_parsed_at":"2023-07-04T21:51:05.161Z","dependency_job_id":null,"html_url":"https://github.com/XuyangBai/KPConv.pytorch","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/XuyangBai%2FKPConv.pytorch","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/XuyangBai%2FKPConv.pytorch/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/XuyangBai%2FKPConv.pytorch/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/XuyangBai%2FKPConv.pytorch/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/XuyangBai","download_url":"https://codeload.github.com/XuyangBai/KPConv.pytorch/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":244571032,"owners_count":20474168,"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-point-clouds","geometric-deep-learning","kpconv","pytorch"],"created_at":"2024-07-31T03:01:12.861Z","updated_at":"2025-03-20T07:33:04.320Z","avatar_url":"https://github.com/XuyangBai.png","language":"C++","funding_links":[],"categories":["C++"],"sub_categories":[],"readme":"## KPConv.pytorch\n\nThis repo is implementation for KPConv(https://arxiv.org/abs/1904.08889) in pytorch.\n\n## TODO\nThere are still some works to be done:\n- [x] Deformable KPConv. Currently I have only implemented the rigid KPConv.\n  - [ ] Regularization loss for the deformable convolution needs to be implemented. I have tried using the deformable convolution layer in part segmention on shapenet without the regularization term, the performance is similar with the rigid convolution counterparts.\n- [x] Speed up. For current implementation, the `collate_fn` where the neighbor indices and pooling indices are calculated, is too slow. In the tf version, the author implement 2 tensroflow C++ wrapper which is quite efficient. I am planing to write C++ extention using pytorch. \n  - [ ] But after I implemented the C++ extention, the evaluation time reduces significantly while the model forward and backward pass still cost about 0.8s per iteration.\n- [ ] Maybe other datasets.\n\n\n## Installation\n\n1. Create an environment from the environment.yml file,\n```\nconda env create -f environment.yml\n```\n2. Compile the customized Tensorflow operators and C++ extension module following the [installation instructions](https://github.com/HuguesTHOMAS/KPConv/blob/master/INSTALL.md) provided by the authors.\n3. Go to `pytorch_ops` dictionary and run `python setup.py install` to build and install the C++ extension for `batch_find_neighbors` function.\n\n\n## Experiments\n\nDue to the time limitation, I have just implemented the experiments on ShapeNet(classification and part segmentation) and ModelNet40. \n\n- Shape Classification on ModelNet40 or ShapeNet.\n```\npython training_ModelNet.py[training_ShapeNetCls.py]\n```\n\n- Part Segmentation on ShapeNet. (I have only implemented the single class part segmentation.)\n```\npython training_ShapeNetPart.py\n```\n\n## Acknowledgment\n\nThank @HuguesTHOMAS for sharing the tensorflow version and valuable explainations. \n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FXuyangBai%2FKPConv.pytorch","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FXuyangBai%2FKPConv.pytorch","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FXuyangBai%2FKPConv.pytorch/lists"}