{"id":13443295,"url":"https://github.com/weiyithu/LiDAR-Distillation","last_synced_at":"2025-03-20T16:30:54.526Z","repository":{"id":48166955,"uuid":"475061619","full_name":"weiyithu/LiDAR-Distillation","owner":"weiyithu","description":"[ECCV 2022] LiDAR Distillation: Bridging the Beam-Induced Domain Gap for 3D Object Detection","archived":false,"fork":false,"pushed_at":"2023-04-26T14:12:47.000Z","size":745,"stargazers_count":106,"open_issues_count":6,"forks_count":10,"subscribers_count":5,"default_branch":"main","last_synced_at":"2024-10-28T06:57:52.320Z","etag":null,"topics":["3d-object-detection","domain-adaptation","lidar-point-cloud"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/weiyithu.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,"publiccode":null,"codemeta":null}},"created_at":"2022-03-28T15:23:21.000Z","updated_at":"2024-10-11T10:12:55.000Z","dependencies_parsed_at":"2024-01-18T14:42:21.783Z","dependency_job_id":"ad316c80-685e-48d5-ba05-903c5455a00a","html_url":"https://github.com/weiyithu/LiDAR-Distillation","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/weiyithu%2FLiDAR-Distillation","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/weiyithu%2FLiDAR-Distillation/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/weiyithu%2FLiDAR-Distillation/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/weiyithu%2FLiDAR-Distillation/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/weiyithu","download_url":"https://codeload.github.com/weiyithu/LiDAR-Distillation/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":244649766,"owners_count":20487487,"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-object-detection","domain-adaptation","lidar-point-cloud"],"created_at":"2024-07-31T03:01:58.773Z","updated_at":"2025-03-20T16:30:54.164Z","avatar_url":"https://github.com/weiyithu.png","language":"Python","funding_links":[],"categories":["Python"],"sub_categories":[],"readme":"# LiDAR Distillation\r\n### [Paper](https://arxiv.org/abs/2203.14956) | [Model](https://cloud.tsinghua.edu.cn/d/8c4541693833471e8b13/) | [中文解读](https://zhuanlan.zhihu.com/p/558773187)\r\n\u003cbr/\u003e\r\n\r\n\u003e [ECCV 2022] LiDAR Distillation: Bridging the Beam-Induced Domain Gap for 3D Object Detection  \r\n\u003e [Yi Wei](https://weiyithu.github.io/), Zibu Wei, [Yongming Rao](https://raoyongming.github.io/), [Jiaxin Li](https://www.jiaxinli.me), [Jiwen Lu](http://ivg.au.tsinghua.edu.cn/Jiwen_Lu/), [Jie Zhou](https://scholar.google.com/citations?user=6a79aPwAAAAJ\u0026hl=en\u0026authuser=1) \r\n\r\n\u003cp align='center'\u003e\r\n\u003cimg src=\"./docs/framework.png\" width='80%'/\u003e\r\n\u003c/p\u003e\r\n\r\n\r\n## Introduction\r\n\r\nIn this paper, we propose the LiDAR Distillation to bridge the domain gap induced by different LiDAR beams for 3D object detection. In many real-world applications, the LiDAR points used by mass-produced robots and vehicles usually have fewer beams than that in large-scale public datasets. Moreover, as the LiDARs are upgraded\r\nto other product models with different beam amount, it becomes challenging to utilize the labeled data captured by previous versions’ high-resolution sensors. Despite the recent progress on domain adaptive 3D\r\ndetection, most methods struggle to eliminate the beam-induced domain gap. \r\n\r\n## Model Zoo\r\n\r\n### Cross-dataset Adaptation\r\n\r\n| model     | method | AP_BEV | AP_3D |  \r\n|------------|----------:|:-------:|:-------:|\r\n| SECOND-IoU | Direct transfer | 32.91  | 17.24 | \r\n| SECOND-IoU | ST3D | 35.92 | 20.19 | \r\n| [SECOND-IoU](https://cloud.tsinghua.edu.cn/f/5a1f958ce2d84e81823b/?dl=1) | Ours | 40.66 | 22.86 | \r\n| [SECOND-IoU](https://cloud.tsinghua.edu.cn/f/a3dd295ac1824d19abf8/?dl=1) | Ours (w / ST3D) | 42.04 | 24.50 |\r\n| PV-RCNN | Direct transfer | 34.50 | 21.47 | \r\n| PV-RCNN | ST3D | 36.42 | 22.99 | \r\n| [PV-RCNN](https://cloud.tsinghua.edu.cn/f/1cea8483c9494df5b46f/?dl=1) | Ours | 43.31 | 25.63 | \r\n| [PV-RCNN](https://cloud.tsinghua.edu.cn/f/3fa7c39d269e4425b4e3/?dl=1) | Ours (w / ST3D) | 44.08 | 26.37 |\r\n| PointPillar | Direct transfer | 27.8 | 12.1 | \r\n| PointPillar | ST3D | 30.6 | 15.6 | \r\n| [PointPillar](https://cloud.tsinghua.edu.cn/f/d919a97bb66d467498d6/?dl=1) | Ours | 40.23 | 19.12| \r\n| [PointPillar](https://cloud.tsinghua.edu.cn/f/8466efb0c91c4d4caeb0/?dl=1) | Ours (w / ST3D) | 40.83 | 20.97 |\r\n\r\nResults of cross-dataset adaptation from Waymo to nuScenes. The training Waymo data used in our work is version 1.0.\r\n\r\n### Single-dataset Adaptation\r\n\r\n|  beams  | method | AP_BEV | AP_3D |  \r\n|------------------|----------:|:-------:|:-------:|\r\n| 32 | Direct transfer | 79.81 | 65.91 | \r\n| 32 | ST3D | 71.29 | 57.57 | \r\n| [32](https://cloud.tsinghua.edu.cn/f/2664ab0e209348a1b5c2/?dl=1) | Ours | 82.22 | 70.15 | \r\n| 32* | Direct transfer | 73.56  | 57.77 | \r\n| 32* | ST3D | 67.08 | 53.30 | \r\n| [32*](https://cloud.tsinghua.edu.cn/f/be087fb6b9634a3c98fe/?dl=1) | Ours | 79.47 | 66.96 | \r\n| 16 | Direct transfer | 64.91  | 47.48 | \r\n| 16 | ST3D | 57.58 | 42.40 | \r\n| [16](https://cloud.tsinghua.edu.cn/f/d8888921ff2a48c9a381/?dl=1) | Ours |  74.32 |  59.87 | \r\n| 16* | Direct transfer |  56.32  | 38.75 | \r\n| 16* | ST3D | 55.63 | 37.02 | \r\n| [16*](https://cloud.tsinghua.edu.cn/f/21d8708a9c464945be61/?dl=1) | Ours | 70.43 | 55.24 | \r\n\r\nResults of single-dataset adaptation on KITTI dataset with PointPillars (moderate difficulty). For SECOND-IoU and PV-RCNN, we find that it is easy to raise cuda error on low-beam data, which is may caused by the bug in spconv. Thus, we do not provide the model but you can still run these experiments with the [yamls](tools/cfgs/kitti_models).\r\n\r\n\r\n## Installation\r\n\r\nPlease refer to [INSTALL.md](docs/INSTALL.md).\r\n\r\n## Getting Started\r\n\r\nPlease refer to [GETTING_STARTED.md](docs/GETTING_STARTED.md).\r\n\r\n## License\r\n\r\nOur code is released under the Apache 2.0 license.\r\n\r\n## Acknowledgement\r\n\r\nOur code is heavily based on [OpenPCDet v0.2](https://github.com/open-mmlab/OpenPCDet/tree/v0.2.0) and [ST3D](https://github.com/CVMI-Lab/ST3D). Thanks OpenPCDet Development Team for their awesome codebase.\r\n\r\n## Citation\r\n\r\nIf you find this project useful in your research, please consider cite:\r\n```\r\n@article{wei2022lidar,\r\n  title={LiDAR Distillation: Bridging the Beam-Induced Domain Gap for 3D Object Detection},\r\n  author={Wei, Yi and Wei, Zibu and Rao, Yongming and Li, Jiaxin and Zhou, Jie and Lu, Jiwen},\r\n  journal={arXiv preprint arXiv:2203.14956},\r\n  year={2022}\r\n}\r\n```\r\n```\r\n@misc{openpcdet2020,\r\n    title={OpenPCDet: An Open-source Toolbox for 3D Object Detection from Point Clouds},\r\n    author={OpenPCDet Development Team},\r\n    howpublished = {\\url{https://github.com/open-mmlab/OpenPCDet}},\r\n    year={2020}\r\n}\r\n```\r\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fweiyithu%2FLiDAR-Distillation","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fweiyithu%2FLiDAR-Distillation","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fweiyithu%2FLiDAR-Distillation/lists"}