{"id":28752983,"url":"https://github.com/filapro/iterdet","last_synced_at":"2025-06-17T00:05:40.717Z","repository":{"id":297420047,"uuid":"996696071","full_name":"filaPro/iterdet","owner":"filaPro","description":"[S+SSPR2020] IterDet: Iterative Scheme for Object Detection in Crowded Environments","archived":false,"fork":false,"pushed_at":"2025-06-05T10:26:54.000Z","size":8401,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"master","last_synced_at":"2025-06-05T11:32:53.622Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mpl-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/filaPro.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,"zenodo":null}},"created_at":"2025-06-05T10:22:08.000Z","updated_at":"2025-06-05T10:26:57.000Z","dependencies_parsed_at":"2025-06-05T11:44:50.279Z","dependency_job_id":null,"html_url":"https://github.com/filaPro/iterdet","commit_stats":null,"previous_names":["filapro/iterdet"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/filaPro/iterdet","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/filaPro%2Fiterdet","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/filaPro%2Fiterdet/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/filaPro%2Fiterdet/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/filaPro%2Fiterdet/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/filaPro","download_url":"https://codeload.github.com/filaPro/iterdet/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/filaPro%2Fiterdet/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":260263608,"owners_count":22982742,"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":[],"created_at":"2025-06-17T00:05:37.573Z","updated_at":"2025-06-17T00:05:40.703Z","avatar_url":"https://github.com/filaPro.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/iterdet-iterative-scheme-for-objectdetection/object-detection-on-crowdhuman-full-body)](https://paperswithcode.com/sota/object-detection-on-crowdhuman-full-body?p=iterdet-iterative-scheme-for-objectdetection)\n[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/iterdet-iterative-scheme-for-objectdetection/object-detection-on-widerperson)](https://paperswithcode.com/sota/object-detection-on-widerperson?p=iterdet-iterative-scheme-for-objectdetection)\n\n# IterDet: Iterative Scheme for Object Detection in Crowded Environments\n\n**News**:\n * :cry: June, 2025. Original repo with [200+](https://web.archive.org/web/20240420000322/github.com/samsunglabs/iterdet) :star: was deleted from `SamsungLabs/iterdet`.\n\nThis project hosts the code for implementing the IterDet scheme for object detection,\nas presented in our paper:\n\n\u003e **IterDet: Iterative Scheme for Object Detection in Crowded Environments**\u003cbr\u003e\n\u003e [Danila Rukhovich](https://github.com/filaPro),\n\u003e [Konstantin Sofiiuk](https://github.com/ksofiyuk),\n\u003e [Danil Galeev](https://github.com/denemmy),\n\u003e [Olga Barinova](https://github.com/OlgaBarinova),\n\u003e [Anton Konushin](https://scholar.google.com/citations?user=ZT_k-wMAAAAJ)\n\u003e \u003cbr\u003e\n\u003e Samsung Research\u003cbr\u003e\n\u003e https://arxiv.org/abs/2005.05708\n\n\u003cp align=\"center\"\u003e\u003cimg src=\"./demo/iterative/scheme.png\" alt=\"drawing\" width=\"90%\"/\u003e\u003c/p\u003e\n\n### Installation\n\nThis implementation is based on [mmdetection](https://github.com/open-mmlab/mmdetection) framework.\n\n\u003cdetails\u003e\n \u003csummary\u003eAll our modifications against their `v2.0.0` release are listed below:\u003c/summary\u003e \n \n * configs/iterative/*\n * demo/iterative/*\n * mmdet/datasets/\\_\\_init\\_\\_.py\n * mmdet/datasets/pipelines/transforms.py\n * mmdet/datasets/pipelines/formating.py\n * mmdet/datasets/crowd_human.py\n * mmdet/models/dense_heads/anchor_head.py\n * mmdet/models/dense_heads/rpn_head.py\n * mmdet/models/roi_heads/bbox_heads/bbox_head.py\n * mmdet/models/backbones/resnet.py\n * mmdet/models/detectors/\\_\\_init\\_\\_.py\n * mmdet/models/detectors/iterdet_faster_rcnn.py\n * mmdet/models/detectors/iderdet_retinanet.py\n * tools/convert_datasets/crowd_human.py\n * tools/convert_datasets/toy.py\n * tools/convert_datasets/wider_person.py\n * requirements/runtime.txt\n * docker/Dockerfile\n \n\u003c/details\u003e\n\nPlease refer to original [install.md](docs/install.md) for installation.\nDo not forget to update the original github repository link, and install [requirements.txt](requirements.txt).\nFor `v1.2.0` release follow `v1` branch.\n\n[Config](configs/iterdet) files and [tools](tools/convert_datasets) \nfor converting annotations to COCO format are provided for the following datasets:\n\n * AdaptIS [ToyV1](https://github.com/saic-vul/adaptis#toyv1-dataset) \n   and [ToyV2](https://github.com/saic-vul/adaptis#toyv2-dataset)\n * [CrowdHuman](https://www.crowdhuman.org/)\n * [WiderPerson](http://www.cbsr.ia.ac.cn/users/sfzhang/WiderPerson/)\n \n### Get Started\n\nPlease see original [getting_started.md](docs/getting_started.md) for the basic usage examples.\nIterdet [configs](configs/iterdet) can be used for [train](tools/dist_train.sh) and [test](tools/dist_test.sh) scripts:\n\n```shell script\nbash tools/dist_train.sh configs/iterdet/crowd_human_full_faster_rcnn_r50_fpn_2x.py 8 --validate\nbash tools/dist_test.sh configs/iterdet/crowd_human_full_faster_rcnn_r50_fpn_2x.py \\\n    work_dirs/iterdet/crowd_human_full_faster_rcnn_r50_fpn_2x/latest.pth 8\n```\n### Models\n\nState-of-the-art models for all datasets are trained on top of Faster RCNN\nbased on ResNet-50. Metrics are given for 2 iterations IterDet inference.\n\n| Dataset              | Download Link                                  | Recall | AP    | mMR   |\n|:--------------------:|:----------------------------------------------:|:------:|:-----:|:-----:|\n| AdaptIS Toy V1       | [toy_v1.pth][toy_v1]                           | 99.60  | 99.25 |       |\n| AdaptIS Toy V2       | [toy_v2.pth][toy_v2]                           | 99.29  | 99.00 |       |\n| CrowdHuman (full)    | [crowd_human_full.pth][crowd_human_full]       | 95.80  | 88.08 | 49.44 |\n| CrowdHuman (visible) | [crowd_human_visible.pth][crowd_human_visible] | 91.63  | 85.33 | 55.61 |\n| WiderPerson          | [wider_person.pth][wider_person]               | 97.15  | 91.95 | 40.78 |\n\n[toy_v1]: https://github.com/saic-vul/iterdet/releases/download/v2.0.0/toy_v1_faster_rcnn_r50_fpn_2x.pth\n[toy_v2]: https://github.com/saic-vul/iterdet/releases/download/v2.0.0/toy_v2_faster_rcnn_r50_fpn_2x.pth\n[crowd_human_full]: https://github.com/saic-vul/iterdet/releases/download/v2.0.0/crowd_human_full_faster_rcnn_r50_fpn_2x.pth\n[crowd_human_visible]: https://github.com/saic-vul/iterdet/releases/download/v2.0.0/crowd_human_visible_faster_rcnn_r50_fpn_2x.pth\n[wider_person]: https://github.com/saic-vul/iterdet/releases/download/v2.0.0/wider_person_faster_rcnn_r50_fpn_2x.pth\n\n### Example Detections\n\n\u003cp align=\"center\"\u003e\u003cimg src=\"./demo/iterative/demo.png\" alt=\"drawing\" width=\"90%\"/\u003e\u003c/p\u003e\nExamples of IterDet results on ToyV1, ToyV2, CrowdHuman (with full body\nannotataions), and WiderPerson. The boxes found on the first and second iterations are\nmarked in green and yellow respectively.\n\n### License\n\nThe code is released under the MPL 2.0 License.\nMPL is a copyleft license that is easy to comply with.\nYou must make the source code for any of your changes available under MPL,\nbut you can combine the MPL software with proprietary code, \nas long as you keep the MPL code in separate files.\n\n### Citation\n\nIf you find this work useful for your research, please cite our paper:\n\n```\n@inproceedings{rukhovich2021iterdet,\n  title={IterDet: Iterative Scheme for Object Detection in Crowded Environments},\n  author={Danila Rukhovich, Konstantin Sofiiuk, Danil Galeev, Olga Barinova, Anton Konushin},\n  booktitle={Structural, Syntactic, and Statistical Pattern Recognition: Joint IAPR International Workshops, S+ SSPR 2020, Padua, Italy, January 21--22, 2021, Proceedings},\n  pages={344},\n  organization={Springer Nature}\n}\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ffilapro%2Fiterdet","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ffilapro%2Fiterdet","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ffilapro%2Fiterdet/lists"}