{"id":20264607,"url":"https://github.com/chrockey/fastpointtransformer-votenet","last_synced_at":"2025-07-10T07:32:40.216Z","repository":{"id":181805282,"uuid":"488551686","full_name":"chrockey/FastPointTransformer-VoteNet","owner":"chrockey","description":"A forked repo from Torch-Points3D for VoteNet with the Fast Point Transformer backbone.","archived":false,"fork":false,"pushed_at":"2022-06-08T10:49:32.000Z","size":56172,"stargazers_count":9,"open_issues_count":0,"forks_count":1,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-04-11T02:51:19.423Z","etag":null,"topics":["3d-vision","computer-vision","cvpr2022","point-cloud","transformer"],"latest_commit_sha":null,"homepage":"https://github.com/POSTECH-CVLab/FastPointTransformer","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/chrockey.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","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":"2022-05-04T10:54:15.000Z","updated_at":"2024-12-20T22:36:40.000Z","dependencies_parsed_at":null,"dependency_job_id":"edce20e7-1195-47a8-830f-ec3787b1aca0","html_url":"https://github.com/chrockey/FastPointTransformer-VoteNet","commit_stats":null,"previous_names":["chrockey/fastpointtransformer-votenet"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/chrockey/FastPointTransformer-VoteNet","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/chrockey%2FFastPointTransformer-VoteNet","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/chrockey%2FFastPointTransformer-VoteNet/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/chrockey%2FFastPointTransformer-VoteNet/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/chrockey%2FFastPointTransformer-VoteNet/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/chrockey","download_url":"https://codeload.github.com/chrockey/FastPointTransformer-VoteNet/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/chrockey%2FFastPointTransformer-VoteNet/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":264545167,"owners_count":23625404,"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-vision","computer-vision","cvpr2022","point-cloud","transformer"],"created_at":"2024-11-14T11:41:49.575Z","updated_at":"2025-07-10T07:32:40.170Z","avatar_url":"https://github.com/chrockey.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Fast Point Transformer\n### [Project Page](http://cvlab.postech.ac.kr/research/FPT/) | [Paper](https://arxiv.org/abs/2112.04702)\nThis repository contains the official code of Fast Point Transformer for 3D object detection experiments below:\n#### 3D object detection with [VoteNet](https://arxiv.org/abs/1904.09664) using [Torch-Points3D](https://github.com/torch-points3d/torch-points3d)\n| Backbone                          | Voxel Size   | mAP@0.25 | mAP@0.5 | Reference |\n|:----------------------------------|:------------:|:--------:|:-------:|:---------:|\n| MinkowskiNet42\u003csup\u003e\u0026dagger;\u003c/sup\u003e | 5cm | 55.3 | 32.8 | [Checkpoint](https://postechackr-my.sharepoint.com/:u:/g/personal/p0125ch_postech_ac_kr/Ecod112ZRnlPp97NNu77N0oBfPtgwzmPxr-tvLvs3eFkwA?download=1) |\n| FastPointTransformer              | 5cm | 59.1 | 35.4 | [Checkpoint](https://postechackr-my.sharepoint.com/:u:/g/personal/p0125ch_postech_ac_kr/EZkpqNi9elVFohE4Xmx66GEBQSjys0ED_h1vUBnJwUz72g?download=1) |\n\n## Installation\nThis repository is developed and tested on\n\n- Ubuntu 20.04\n- Conda 4.12.0\n- CUDA 11.1\n- Python 3.8.13\n- PyTorch 1.7.1\n- MinkowskiEngine 0.5.4\n\n### Environment Setup\nSince this repo is forked from the [Torch-Points3D repo](https://github.com/torch-points3d/torch-points3d), you can setup the environment by following the the [Torch-Points3D repo](https://github.com/torch-points3d/torch-points3d).\nWe also provide a docker image to ease the environment setup.\nYou can pull and run the docker image via the following commands:\n```bash\n~$ docker pull chrockey/fpt-votenet:v0.1.0\n~$ docker run {docker_arguments} chrockey/fpt-votenet:v0.1.0 # interactive mode\n```\nWithin the docker container, you may find a conda environment named `tp3d-fpt`:\n```bash\n~$ conda activate tp3d-fpt\n(tp3d-fpt) ~$ python -c \"import torch; import cuda_sparse_ops\"\n```\n\n### Dataset Preparation\nFirst, you need to make a symbolic link for raw ScanNet V2 dataset via the following command:\n```bash\n~/FastPointTransformer-VoteNet$ ln -s {dir_to_scannet_v2_dataset} data/scannet-sparse/raw\n```\nAnd then, your data directory should look like the structure below:\n```\n~/FastPointTransformer-VoteNet/data/scannet-sparse\n└── raw\n    ├── metadata\n    ├── scans\n    ├── scans_test\n    └── scannetv2-labels.combined.tsv\n```\n\n## Training \u0026 Evaluation\nAfter linking the raw dataset, run the provided training script (`train_scripts/train_votenet_fpt.sh`).\nThe training outputs will be saved in the `outputs` directory.\n```bash\n~/FastPointTransformer-VoteNet$ conda activate tp3d-fpt\n(tp3d-fpt) ~/FastPointTransformer-VoteNet$ sh train_scripts/train_votenet_fpt.sh\n```\nAnd then, you can evaluate the model as:\n```bash\n(tp3d-fpt) ~/FastPointTransformer-VoteNet$ sh eval_scripts/eval_votenet_fpt.sh\n```\nNote that you may need to modify the checkpoint directory within the script (`eval_scripts/eval_votenet_fpt.sh`).\n\n## LICENSE\nFor [Torch-Points3D](https://github.com/torch-points3d/torch-points3d) repo, please check [the license](https://github.com/chrockey/FastPointTransformer-VoteNet/blob/main/LICENSE).\n\n## Acknowledgement\n\nThis repo is forked from [Torch-Points3D](https://github.com/torch-points3d/torch-points3d) repo.\nIf you use our model, please consider citing [Torch-Points3D](https://github.com/torch-points3d/torch-points3d) and [VoteNet](https://arxiv.org/abs/1904.09664) as well.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fchrockey%2Ffastpointtransformer-votenet","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fchrockey%2Ffastpointtransformer-votenet","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fchrockey%2Ffastpointtransformer-votenet/lists"}