{"id":28136982,"url":"https://github.com/res2net/res2net-detectron2","last_synced_at":"2025-10-18T01:19:12.713Z","repository":{"id":182996374,"uuid":"262964774","full_name":"Res2Net/Res2Net-detectron2","owner":"Res2Net","description":"Res2Net for Panoptic Segmentation based on detectron2 (SOTA results).","archived":false,"fork":false,"pushed_at":"2020-06-08T02:13:16.000Z","size":645,"stargazers_count":30,"open_issues_count":2,"forks_count":11,"subscribers_count":3,"default_branch":"master","last_synced_at":"2024-04-23T00:23:22.815Z","etag":null,"topics":["detectron2","multi-scale","panoptic","res2net","segmentation","sota"],"latest_commit_sha":null,"homepage":"https://mmcheng.net/res2net/","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/Res2Net.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":".github/CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":".github/CODE_OF_CONDUCT.md","threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null}},"created_at":"2020-05-11T07:03:23.000Z","updated_at":"2022-07-28T10:58:21.000Z","dependencies_parsed_at":"2023-07-22T10:49:02.440Z","dependency_job_id":null,"html_url":"https://github.com/Res2Net/Res2Net-detectron2","commit_stats":null,"previous_names":["res2net/res2net-detectron2"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Res2Net%2FRes2Net-detectron2","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Res2Net%2FRes2Net-detectron2/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Res2Net%2FRes2Net-detectron2/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Res2Net%2FRes2Net-detectron2/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Res2Net","download_url":"https://codeload.github.com/Res2Net/Res2Net-detectron2/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":254179885,"owners_count":22027884,"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":["detectron2","multi-scale","panoptic","res2net","segmentation","sota"],"created_at":"2025-05-14T16:20:48.678Z","updated_at":"2025-10-18T01:19:12.597Z","avatar_url":"https://github.com/Res2Net.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Res2Net for Panoptic Segmentation based on detectron2.\n\n## Introduction\n\nWe propose a novel building block for CNNs, namely Res2Net, by constructing hierarchical residual-like connections within one single residual block. The Res2Net represents multi-scale features at a granular level and increases the range of receptive fields for each network layer.\n\n## Performance on COCO dataset\n\u003ctable\u003e\u003ctbody\u003e\n\u003c!-- START TABLE --\u003e\n\u003c!-- TABLE HEADER --\u003e\n\u003cth valign=\"bottom\"\u003eName\u003c/th\u003e\n\u003cth valign=\"bottom\"\u003elr\u003cbr/\u003esched\u003c/th\u003e\n\u003cth valign=\"bottom\"\u003etrain\u003cbr/\u003emem\u003cbr/\u003e(GB)\u003c/th\u003e\n\u003cth valign=\"bottom\"\u003ebox\u003cbr/\u003eAP\u003c/th\u003e\n\u003cth valign=\"bottom\"\u003emask\u003cbr/\u003eAP\u003c/th\u003e\n\u003cth valign=\"bottom\"\u003ePQ\u003c/th\u003e\n\u003cth valign=\"bottom\"\u003edownload\u003c/th\u003e\n\u003c!-- TABLE BODY --\u003e\n\u003c!-- ROW: panoptic_fpn_R_50_1x --\u003e\n \u003ctr\u003e\u003ctd align=\"left\"\u003e\u003ca href=\"configs/COCO-PanopticSegmentation/panoptic_fpn_R_50_1x.yaml\"\u003eR50-FPN\u003c/a\u003e\u003c/td\u003e\n\u003ctd align=\"center\"\u003e1x\u003c/td\u003e\n\u003ctd align=\"center\"\u003e4.8\u003c/td\u003e\n\u003ctd align=\"center\"\u003e37.6\u003c/td\u003e\n\u003ctd align=\"center\"\u003e34.7\u003c/td\u003e\n\u003ctd align=\"center\"\u003e39.4\u003c/td\u003e\n\u003ctd align=\"center\"\u003e\u003ca href=\"https://dl.fbaipublicfiles.com/detectron2/COCO-PanopticSegmentation/panoptic_fpn_R_50_1x/139514544/model_final_dbfeb4.pkl\"\u003emodel\u003c/a\u003e\u0026nbsp;|\u0026nbsp;\u003ca href=\"https://dl.fbaipublicfiles.com/detectron2/COCO-PanopticSegmentation/panoptic_fpn_R_50_1x/139514544/metrics.json\"\u003emetrics\u003c/a\u003e\u003c/td\u003e\n\u003c/tr\u003e\n\u003c!-- ROW: panoptic_fpn_R_50_3x --\u003e\n \u003ctr\u003e\u003ctd align=\"left\"\u003e\u003ca href=\"configs/COCO-PanopticSegmentation/panoptic_fpn_R_50_3x.yaml\"\u003eR50-FPN\u003c/a\u003e\u003c/td\u003e\n\u003ctd align=\"center\"\u003e3x\u003c/td\u003e\n\u003ctd align=\"center\"\u003e4.8\u003c/td\u003e\n\u003ctd align=\"center\"\u003e40.0\u003c/td\u003e\n\u003ctd align=\"center\"\u003e36.5\u003c/td\u003e\n\u003ctd align=\"center\"\u003e41.5\u003c/td\u003e\n\u003ctd align=\"center\"\u003e\u003ca href=\"https://dl.fbaipublicfiles.com/detectron2/COCO-PanopticSegmentation/panoptic_fpn_R_50_3x/139514569/model_final_c10459.pkl\"\u003emodel\u003c/a\u003e\u0026nbsp;|\u0026nbsp;\u003ca href=\"https://dl.fbaipublicfiles.com/detectron2/COCO-PanopticSegmentation/panoptic_fpn_R_50_3x/139514569/metrics.json\"\u003emetrics\u003c/a\u003e\u003c/td\u003e\n\u003c/tr\u003e\n\u003c!-- ROW: panoptic_fpn_R_101_3x --\u003e\n \u003ctr\u003e\u003ctd align=\"left\"\u003e\u003ca href=\"configs/COCO-PanopticSegmentation/panoptic_fpn_R_101_3x.yaml\"\u003eR101-FPN\u003c/a\u003e\u003c/td\u003e\n\u003ctd align=\"center\"\u003e3x\u003c/td\u003e\n\u003ctd align=\"center\"\u003e6.0\u003c/td\u003e\n\u003ctd align=\"center\"\u003e42.4\u003c/td\u003e\n\u003ctd align=\"center\"\u003e38.5\u003c/td\u003e\n\u003ctd align=\"center\"\u003e43.0\u003c/td\u003e\n\u003ctd align=\"center\"\u003e\u003ca href=\"https://dl.fbaipublicfiles.com/detectron2/COCO-PanopticSegmentation/panoptic_fpn_R_101_3x/139514519/model_final_cafdb1.pkl\"\u003emodel\u003c/a\u003e\u0026nbsp;|\u0026nbsp;\u003ca href=\"https://dl.fbaipublicfiles.com/detectron2/COCO-PanopticSegmentation/panoptic_fpn_R_101_3x/139514519/metrics.json\"\u003emetrics\u003c/a\u003e\u003c/td\u003e\n\u003c/tr\u003e\n\u003c!-- ROW: panoptic_fpn_R2_101_3x --\u003e\n \u003ctr\u003e\u003ctd align=\"left\"\u003e\u003ca href=\"configs/COCO-PanopticSegmentation/panoptic_fpn_R2_101_3x.yaml\"\u003eRes2Net101-FPN\u003c/a\u003e\u003c/td\u003e\n\u003ctd align=\"center\"\u003e3x\u003c/td\u003e\n\u003ctd align=\"center\"\u003e6.0\u003c/td\u003e\n\u003ctd align=\"center\"\u003e44.0\u003c/td\u003e\n\u003ctd align=\"center\"\u003e39.6\u003c/td\u003e\n\u003ctd align=\"center\"\u003e44.5\u003c/td\u003e\n\u003ctd align=\"center\"\u003e\u003ca href=\"https://mailnankaieducn-my.sharepoint.com/:u:/g/personal/shgao_mail_nankai_edu_cn/EU024RDiIxtJs2xz2zl_7bkBRXiPFcRukFLcB4gVYxzasw?e=PHU5jk\"\u003emodel\u003c/a\u003e\u0026nbsp;|\u0026nbsp;\u003ca href=\"results/panoptic_seg_res2net101_fpn_x3.txt\"\u003emetrics\u003c/a\u003e\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\u003c/table\u003e\n\n- Res2Net101 has the similar parameters and FLOPs compared with ResNet101.\n- We only test and show results of Panoptic Segmentation on detectron2, the detection and instance segmentation on detectron2 is also supported in this repo.\n- Res2Net ImageNet pretrained models are in [Res2Net-PretrainedModels](https://github.com/Res2Net/Res2Net-PretrainedModels).\n- More applications of Res2Net are in [Res2Net-Github](https://github.com/Res2Net/).\n\n## Usage\n- Use the tools/convert-torchvision-to-d2.py to transfer the ImageNet pretrained model of Res2Net101 to detectron2 supported format. Or just download the converted model from\nthis [link](https://mailnankaieducn-my.sharepoint.com/:u:/g/personal/shgao_mail_nankai_edu_cn/EZhtWgMRlxpGtJmtJ2zP1_QBqvmu_FJ05vUgOq30ElT9yg?e=e5FRD8).\n- Use the command to train:\n```\n./tools/train_net.py --num-gpus 8 --config-file configs/COCO-PanopticSegmentation/panoptic_fpn_R2_101_3x.yaml\n```\n\n## Citation\nIf you find this work or code is helpful in your research, please cite:\n```\n@article{gao2019res2net,\n  title={Res2Net: A New Multi-scale Backbone Architecture},\n  author={Gao, Shang-Hua and Cheng, Ming-Ming and Zhao, Kai and Zhang, Xin-Yu and Yang, Ming-Hsuan and Torr, Philip},\n  journal={IEEE TPAMI},\n  year={2020},\n  doi={10.1109/TPAMI.2019.2938758}, \n}\n```\n\nFor more details of detectron2, please refer to\nthe detectron2 repo.\n\n\u003cimg src=\".github/Detectron2-Logo-Horz.svg\" width=\"300\" \u003e\n\nDetectron2 is Facebook AI Research's next generation software system\nthat implements state-of-the-art object detection algorithms.\nIt is a ground-up rewrite of the previous version,\n[Detectron](https://github.com/facebookresearch/Detectron/),\nand it originates from [maskrcnn-benchmark](https://github.com/facebookresearch/maskrcnn-benchmark/).\n\n\u003cdiv align=\"center\"\u003e\n  \u003cimg src=\"https://user-images.githubusercontent.com/1381301/66535560-d3422200-eace-11e9-9123-5535d469db19.png\"/\u003e\n\u003c/div\u003e\n\n### What's New\n* It is powered by the [PyTorch](https://pytorch.org) deep learning framework.\n* Includes more features such as panoptic segmentation, densepose, Cascade R-CNN, rotated bounding boxes, etc.\n* Can be used as a library to support [different projects](projects/) on top of it.\n  We'll open source more research projects in this way.\n* It [trains much faster](https://detectron2.readthedocs.io/notes/benchmarks.html).\n\nSee our [blog post](https://ai.facebook.com/blog/-detectron2-a-pytorch-based-modular-object-detection-library-/)\nto see more demos and learn about detectron2.\n\n## Installation\n\nSee [INSTALL.md](INSTALL.md).\n\n## Quick Start\n\nSee [GETTING_STARTED.md](GETTING_STARTED.md),\nor the [Colab Notebook](https://colab.research.google.com/drive/16jcaJoc6bCFAQ96jDe2HwtXj7BMD_-m5).\n\nLearn more at our [documentation](https://detectron2.readthedocs.org).\nAnd see [projects/](projects/) for some projects that are built on top of detectron2.\n\n## Model Zoo and Baselines\n\nWe provide a large set of baseline results and trained models available for download in the [Detectron2 Model Zoo](MODEL_ZOO.md).\n\n\n## License\n\nDetectron2 is released under the [Apache 2.0 license](LICENSE).\n\n## Citing Detectron2\n\nIf you use Detectron2 in your research or wish to refer to the baseline results published in the [Model Zoo](MODEL_ZOO.md), please use the following BibTeX entry.\n\n```BibTeX\n@misc{wu2019detectron2,\n  author =       {Yuxin Wu and Alexander Kirillov and Francisco Massa and\n                  Wan-Yen Lo and Ross Girshick},\n  title =        {Detectron2},\n  howpublished = {\\url{https://github.com/facebookresearch/detectron2}},\n  year =         {2019}\n}\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fres2net%2Fres2net-detectron2","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fres2net%2Fres2net-detectron2","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fres2net%2Fres2net-detectron2/lists"}