{"id":13704780,"url":"https://github.com/yihongXU/TransCenter","last_synced_at":"2025-05-05T12:32:30.884Z","repository":{"id":42053786,"uuid":"355262265","full_name":"yihongXU/TransCenter","owner":"yihongXU","description":"This is the official implementation of TransCenter (TPAMI). The code  and pretrained models are now available here: https://gitlab.inria.fr/yixu/TransCenter_official.","archived":false,"fork":false,"pushed_at":"2023-08-03T09:23:55.000Z","size":36293,"stargazers_count":107,"open_issues_count":11,"forks_count":7,"subscribers_count":10,"default_branch":"main","last_synced_at":"2024-08-03T22:14:03.603Z","etag":null,"topics":["computer-vision","deep-learning","multiple-object-tracking","pytorch","transformers"],"latest_commit_sha":null,"homepage":"https://team.inria.fr/robotlearn/transcenter-transformers-with-dense-queriesfor-multiple-object-tracking/","language":null,"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/yihongXU.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":"2021-04-06T16:43:32.000Z","updated_at":"2024-06-03T08:00:16.000Z","dependencies_parsed_at":"2022-09-18T12:50:29.350Z","dependency_job_id":"a790973a-71b3-4e68-957e-27f0480e2977","html_url":"https://github.com/yihongXU/TransCenter","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/yihongXU%2FTransCenter","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/yihongXU%2FTransCenter/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/yihongXU%2FTransCenter/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/yihongXU%2FTransCenter/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/yihongXU","download_url":"https://codeload.github.com/yihongXU/TransCenter/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":224448712,"owners_count":17313105,"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":["computer-vision","deep-learning","multiple-object-tracking","pytorch","transformers"],"created_at":"2024-08-02T22:00:16.582Z","updated_at":"2024-11-13T12:31:03.274Z","avatar_url":"https://github.com/yihongXU.png","language":null,"funding_links":[],"categories":["其他_机器视觉","算法论文"],"sub_categories":["网络服务_其他","**2021**"],"readme":"## TransCenter: Transformers with Dense Representations for Multiple-Object Tracking \u003cbr /\u003e\n## The work is accepted for TPAMI 2022.\n## An update towards a more efficient and powerful TransCenter, TransCenter-Lite! ##\n\n## The code for TransCenter and TransCenter-Lite is now available, you can find the code and pretrained models at https://gitlab.inria.fr/robotlearn/TransCenter_official.\n\n**TransCenter: Transformers with Dense Representations for Multiple-Object Tracking** \u003cbr /\u003e\n[Yihong Xu](https://team.inria.fr/robotlearn/team-members/yihong-xu/), [Yutong Ban](https://people.csail.mit.edu/yban/index.html), [Guillaume Delorme](https://team.inria.fr/robotlearn/team-members/guillaume-delorme/), [Chuang Gan](https://people.csail.mit.edu/ganchuang/), [Daniela Rus](http://danielarus.csail.mit.edu/), [Xavier Alameda-Pineda](http://xavirema.eu/) \u003cbr /\u003e\n**[[Paper](https://arxiv.org/abs/2103.15145)]** **[[Project](https://team.inria.fr/robotlearn/transcenter-transformers-with-dense-queriesfor-multiple-object-tracking/)]**\u003cbr /\u003e\n\n\u003cdiv align=\"center\"\u003e\n  \u003cimg src=\"https://github.com/yihongXU/TransCenter/raw/main/eTransCenter_pipeline.png\" width=\"1200px\" /\u003e\n\u003c/div\u003e\n\n\u003cbr /\u003e\u003cbr /\u003e\n**MOT20 example:** \u003cbr /\u003e\n![](https://github.com/yihongXU/TransCenter/blob/main/transcenter_mot20_example.gif)\n\n\n## Bibtex\n**If you find this code useful, please star the project and consider citing:** \u003cbr /\u003e\n```\n@misc{xu2021transcenter,\n      title={TransCenter: Transformers with Dense Representations for Multiple-Object Tracking}, \n      author={Yihong Xu and Yutong Ban and Guillaume Delorme and Chuang Gan and Daniela Rus and Xavier Alameda-Pineda},\n      year={2021},\n      eprint={2103.15145},\n      archivePrefix={arXiv},\n      primaryClass={cs.CV}\n}\n```\n\n\n## MOTChallenge Results\n\n***For TransCenter***:\n\nMOT17 public detections:\n     \n| Pretrained| MOTA     | MOTP     | IDF1 |  FP    | FN    | IDS |\n|-----------|----------|----------|--------|-------|------|----------------|\n|   CoCo  |  71.9%   |  80.5%   | 64.1% | 27,356  | 126,860  |     4,118     |\n|   CH    |  75.9%   |  81.2%   | 65.9%  | 30,190 | 100,999 |     4,626    |\n\nMOT20 public detections:\n   \n| Pretrained| MOTA     | MOTP     | IDF1 |  FP    | FN    | IDS |\n|-----------|----------|----------|--------|-------|------|----------------|\n|   CoCo    |  67.7%   |  79.8%   | 58.9%  | 54,967   | 108,376  |     3,707     |\n|   CH      |  72.8%   |  81.0%   | 57.6%  | 28,026  | 110,312  |     2,621     |\n\n\nMOT17 private detections:\n   \n| Pretrained| MOTA     | MOTP     | IDF1 |  FP    | FN    | IDS |\n|-----------|----------|----------|--------|-------|------|----------------|\n|   CoCo  |  72.7%   |  80.3%   | 64.0% | 33,807   | 115,542  |    4,719     |\n|   CH    |  76.2%   |  81.1%   | 65.5% | 40,101 | 88,827 |     5,394    |\n\nMOT20 private detections:\n\n| Pretrained| MOTA     | MOTP     | IDF1 |  FP    | FN    | IDS |\n|-----------|----------|----------|--------|-------|------|----------------|\n|   CoCo   |  67.7%   |  79.8%   | 58.7% | 56,435  | 107,163 |     3,759     |\n|   CH   |  72.9%   |  81.0%   | 57.7%  | 28,596  | 108,982  |     2,625     |\n\n\n**Note:** \n- The results can be slightly different depending on the running environment.\n- We might keep updating the results in the near future.\n\n## Acknowledgement\n\nThe code for TransCenterV2, TransCenter-Lite is modified and network pre-trained weights are obtained from the following repositories:\n\n1) The PVTv2 backbone pretrained models from PVTv2.\n2) The data format conversion code is modified from CenterTrack.\n\n[**CenterTrack**](https://github.com/xingyizhou/CenterTrack), [**Deformable-DETR**](https://github.com/fundamentalvision/Deformable-DETR), [**Tracktor**](https://github.com/phil-bergmann/tracking_wo_bnw).\n```\n@article{zhou2020tracking,\n  title={Tracking Objects as Points},\n  author={Zhou, Xingyi and Koltun, Vladlen and Kr{\\\"a}henb{\\\"u}hl, Philipp},\n  journal={ECCV},\n  year={2020}\n}\n\n@InProceedings{tracktor_2019_ICCV,\nauthor = {Bergmann, Philipp and Meinhardt, Tim and Leal{-}Taix{\\'{e}}, Laura},\ntitle = {Tracking Without Bells and Whistles},\nbooktitle = {The IEEE International Conference on Computer Vision (ICCV)},\nmonth = {October},\nyear = {2019}}\n\n@article{zhu2020deformable,\n  title={Deformable DETR: Deformable Transformers for End-to-End Object Detection},\n  author={Zhu, Xizhou and Su, Weijie and Lu, Lewei and Li, Bin and Wang, Xiaogang and Dai, Jifeng},\n  journal={arXiv preprint arXiv:2010.04159},\n  year={2020}\n}\n\n@article{zhang2021bytetrack,\n  title={ByteTrack: Multi-Object Tracking by Associating Every Detection Box},\n  author={Zhang, Yifu and Sun, Peize and Jiang, Yi and Yu, Dongdong and Yuan, Zehuan and Luo, Ping and Liu, Wenyu and Wang, Xinggang},\n  journal={arXiv preprint arXiv:2110.06864},\n  year={2021}\n}\n\n@article{wang2021pvtv2,\n  title={Pvtv2: Improved baselines with pyramid vision transformer},\n  author={Wang, Wenhai and Xie, Enze and Li, Xiang and Fan, Deng-Ping and Song, Kaitao and Liang, Ding and Lu, Tong and Luo, Ping and Shao, Ling},\n  journal={Computational Visual Media},\n  volume={8},\n  number={3},\n  pages={1--10},\n  year={2022},\n  publisher={Springer}\n}\n```\nSeveral modules are from:\n\n**MOT Metrics in Python**: [**py-motmetrics**](https://github.com/cheind/py-motmetrics)\n\n**Soft-NMS**: [**Soft-NMS**](https://github.com/DocF/Soft-NMS)\n\n**DETR**: [**DETR**](https://github.com/facebookresearch/detr)\n\n**DCNv2**: [**DCNv2**](https://github.com/CharlesShang/DCNv2)\n\n**PVTv2**: [**PVTv2**](https://github.com/whai362/PVT)\n\n**ByteTrack**: [**ByteTrack**](https://github.com/ifzhang/ByteTrack)\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FyihongXU%2FTransCenter","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FyihongXU%2FTransCenter","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FyihongXU%2FTransCenter/lists"}