{"id":15027994,"url":"https://github.com/res2net/res2net-pretrainedmodels","last_synced_at":"2025-05-16T14:04:53.702Z","repository":{"id":48168920,"uuid":"195163550","full_name":"Res2Net/Res2Net-PretrainedModels","owner":"Res2Net","description":"(ImageNet pretrained models) The official pytorch implemention of the TPAMI paper \"Res2Net: A New Multi-scale Backbone Architecture\"","archived":false,"fork":false,"pushed_at":"2022-12-08T12:54:57.000Z","size":78,"stargazers_count":1094,"open_issues_count":10,"forks_count":216,"subscribers_count":28,"default_branch":"master","last_synced_at":"2025-05-16T14:04:45.416Z","etag":null,"topics":["backbone","jittor","multi-scale","pytorch","res2net"],"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":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Res2Net.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2019-07-04T03:35:43.000Z","updated_at":"2025-05-14T11:38:26.000Z","dependencies_parsed_at":"2023-01-25T13:45:16.112Z","dependency_job_id":null,"html_url":"https://github.com/Res2Net/Res2Net-PretrainedModels","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/Res2Net%2FRes2Net-PretrainedModels","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Res2Net%2FRes2Net-PretrainedModels/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Res2Net%2FRes2Net-PretrainedModels/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Res2Net%2FRes2Net-PretrainedModels/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Res2Net","download_url":"https://codeload.github.com/Res2Net/Res2Net-PretrainedModels/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":254544146,"owners_count":22088807,"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":["backbone","jittor","multi-scale","pytorch","res2net"],"created_at":"2024-09-24T20:07:26.326Z","updated_at":"2025-05-16T14:04:53.646Z","avatar_url":"https://github.com/Res2Net.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Res2Net\nThe official pytorch implemention of the paper [\"Res2Net: A New Multi-scale Backbone Architecture\"](https://arxiv.org/pdf/1904.01169.pdf)\n\nOur paper is accepted by **IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)**.\n## Update\n- 2020.10.20 PaddlePaddle version Res2Net achieves 85.13% top-1 acc. on ImageNet: [PaddlePaddle Res2Net](https://github.com/PaddlePaddle/PaddleClas/blob/master/docs/en/advanced_tutorials/distillation/distillation_en.md).\n- 2020.8.21 Online demo for detection and segmentation using Res2Net is released: http://mc.nankai.edu.cn/res2net-det\n- 2020.7.29 The training code of Res2Net on ImageNet is released https://github.com/Res2Net/Res2Net-ImageNet-Training (non-commercial use only)\n- 2020.6.1 Res2Net is now in the official model zoo of the new deep learning framework [**Jittor**](https://github.com/Jittor/jittor).\n- 2020.5.21 Res2Net is now one of the basic bonebones in MMDetection v2 framework https://github.com/open-mmlab/mmdetection.\nUsing MMDetection v2 with Res2Net achieves better performance with less computational cost.\n- 2020.5.11 Res2Net achieves about 2% performance gain on Panoptic Segmentation based on detectron2 with no trick. We have released our code on: https://github.com/Res2Net/Res2Net-detectron2.\n- 2020.2.24 Our Res2Net_v1b achieves a considerable performance gain on mmdetection compared with existing backbone models.\nWe have released our code on: https://github.com/Res2Net/mmdetection. Detailed comparision between our method and HRNet, which previously generates best results, could be found at: https://github.com/Res2Net/mmdetection/tree/master/configs/res2net\n- 2020.2.21: Pretrained models of Res2Net_v1b with more than 2% improvement on ImageNet top1 acc. compared with original version of Res2Net are released! Res2Net_v1b achieves much better performance when transfer to other tasks such as object detection and semantic segmentation.\n## Introduction\nWe propose a novel building block for CNNs, namely Res2Net, by constructing hierarchical residual-like\nconnections within one single residual block. The Res2Net represents multi-scale features at a granular level and increases the range\nof receptive fields for each network layer. The proposed Res2Net block can be plugged into the state-of-the-art backbone CNN models,\ne.g. , ResNet, ResNeXt, BigLittleNet, and DLA. We evaluate the Res2Net block on all these models and demonstrate consistent performance gains over baseline models.\n\u003cp align=\"center\"\u003e\n\t\u003cimg src=\"https://mftp.mmcheng.net/imgs800/19Res2Net.jpg\" alt=\"Sample\"  width=\"500\"\u003e\n\t\u003cp align=\"center\"\u003e\n\t\t\u003cem\u003eRes2Net module\u003c/em\u003e\n\t\u003c/p\u003e\n\u003c/p\u003e\n\n## Useage\n### Requirement\nPyTorch\u003e=0.4.1\n### Examples \n```\ngit clone https://github.com/gasvn/Res2Net.git\n\nfrom res2net import res2net50\nmodel = res2net50(pretrained=True)\n\n```\nInput image should be normalized as follows:\n```\nnormalize = transforms.Normalize(mean=[0.485, 0.456, 0.406],\n                                  std=[0.229, 0.224, 0.225])\n```\n(By default, the model will be downloaded automatically.\nIf the default download link is not available, please refer to the Download Link listed on **Pretrained models**.)\n## Pretrained models\n| model |#Params | MACCs |top-1 error| top-5 error| Link |\n| :--: | :--: | :--: | :--: | :--: | :--: |\n| Res2Net-50-48w-2s  | 25.29M | 4.2 | 22.68 | 6.47 |[OneDrive](https://1drv.ms/u/s!AkxDDnOtroRPbo7RnRUz-7ejhLg?e=gU2EZG)\n| Res2Net-50-26w-4s  | 25.70M | 4.2 | 22.01 | 6.15 |[OneDrive](https://1drv.ms/u/s!AkxDDnOtroRPbMavn7eawKhvCPY?e=TBHOuT)\n| Res2Net-50-14w-8s  | 25.06M | 4.2 | 21.86 | 6.14 |[OneDrive](https://1drv.ms/u/s!AkxDDnOtroRPdOTqhF8ne_aakDI?e=EVb8Ri)\n| Res2Net-50-26w-6s  | 37.05M | 6.3 | 21.42 | 5.87 |[OneDrive](https://1drv.ms/u/s!AkxDDnOtroRPc2mqy1h8324sxxI?e=Go4p7I)\n| Res2Net-50-26w-8s  | 48.40M | 8.3 | 20.80 | 5.63 |[OneDrive](https://1drv.ms/u/s!AkxDDnOtroRPdTrAd_Afzc26Z7Q?e=slYqsR)\n| Res2Net-101-26w-4s | 45.21M | 8.1 | 20.81 | 5.57 |[OneDrive](https://1drv.ms/u/s!AkxDDnOtroRPcJRgTLkahL0cFYw?e=nwbnic)\n| Res2NeXt-50        | 24.67M | 4.2 | 21.76 | 6.09 |[OneDrive](https://1drv.ms/u/s!AkxDDnOtroRPcWlWLXBuKxma7DQ?e=mt4dQf)\n| Res2Net-DLA-60     | 21.15M | 4.2 | 21.53 | 5.80 |[OneDrive](https://1drv.ms/u/s!AkxDDnOtroRPbWAqdcatece24vs?e=t3shXH)\n| Res2NeXt-DLA-60    | 17.33M | 3.6 | 21.55 | 5.86 |[OneDrive](https://1drv.ms/u/s!AkxDDnOtroRPcjxCM0kAYHEaEd0?e=9WrBpj)\n| **Res2Net-v1b-50** | 25.72M | 4.5 | 19.73 | 4.96 |[Link](https://shanghuagao.oss-cn-beijing.aliyuncs.com/res2net/res2net50_v1b_26w_4s-3cf99910.pth)\n| **Res2Net-v1b-101**| 45.23M | 8.3 | 18.77 | 4.64 |[Link](https://shanghuagao.oss-cn-beijing.aliyuncs.com/res2net/res2net101_v1b_26w_4s-0812c246.pth)\n| **Res2Net-v1d-200-SSLD**| 76.21M | 15.7 | 14.87 | 2.58 |[PaddlePaddleLink](https://paddle-imagenet-models-name.bj.bcebos.com/Res2Net200_vd_26w_4s_ssld_pretrained.tar)\n\n#### News\n- Res2Net_v1b is now available.\n- You can load the pretrained model by using `pretrained = True`.\n\nThe download link from Baidu Disk is now available. ([Baidu Disk](https://pan.baidu.com/s/1BP7X222ZPqOndbojwOPjkw) password: **vbix**)\n## Applications\nOther applications such as Classification, Instance segmentation, Object detection, Semantic segmentation, Salient object detection, Class activation map,Tumor segmentation on CT scans can be found on https://mmcheng.net/res2net/ .\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={2021},\n  doi={10.1109/TPAMI.2019.2938758}, \n}\n```\n## Contact\nIf you have any questions, feel free to E-mail me via: `shgao(at)live.com`\n\n## License\nThe code is released under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International Public License for Noncommercial use only. Any commercial use should get formal permission first.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fres2net%2Fres2net-pretrainedmodels","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fres2net%2Fres2net-pretrainedmodels","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fres2net%2Fres2net-pretrainedmodels/lists"}