{"id":13546338,"url":"https://github.com/ZHKKKe/PixelSSL","last_synced_at":"2025-04-02T18:30:36.106Z","repository":{"id":44446183,"uuid":"287060497","full_name":"ZHKKKe/PixelSSL","owner":"ZHKKKe","description":"A PyTorch-based Semi-Supervised Learning (SSL) Codebase for Pixel-wise (Pixel) Vision Tasks [ECCV 2020]","archived":false,"fork":false,"pushed_at":"2021-04-06T15:10:13.000Z","size":322,"stargazers_count":290,"open_issues_count":5,"forks_count":30,"subscribers_count":13,"default_branch":"master","last_synced_at":"2025-03-24T09:34:38.864Z","etag":null,"topics":["computer-vision","pixel-wise-task","pytorch-implementation","semi-supervised","toolbox"],"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/ZHKKKe.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}},"created_at":"2020-08-12T16:16:45.000Z","updated_at":"2025-03-14T03:28:25.000Z","dependencies_parsed_at":"2022-09-06T04:01:26.425Z","dependency_job_id":null,"html_url":"https://github.com/ZHKKKe/PixelSSL","commit_stats":null,"previous_names":[],"tags_count":5,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ZHKKKe%2FPixelSSL","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ZHKKKe%2FPixelSSL/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ZHKKKe%2FPixelSSL/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ZHKKKe%2FPixelSSL/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/ZHKKKe","download_url":"https://codeload.github.com/ZHKKKe/PixelSSL/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":246869599,"owners_count":20847157,"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","pixel-wise-task","pytorch-implementation","semi-supervised","toolbox"],"created_at":"2024-08-01T12:00:35.667Z","updated_at":"2025-04-02T18:30:35.589Z","avatar_url":"https://github.com/ZHKKKe.png","language":"Python","readme":"\u003cdiv align=\"center\"\u003e\n  \u003cimg src=\"docs/img/pixelssl-logo.png\" width=\"650\"/\u003e\n\u003c/div\u003e\n\n---\n\nPixelSSL is a PyTorch-based semi-supervised learning (SSL) codebase for pixel-wise (Pixel) vision tasks.\n\nThe purpose of this project is to promote the research and application of semi-supervised learning on pixel-wise vision tasks. PixelSSL provides two major features:\n- Interface for implementing new semi-supervised algorithms\n- Template for encapsulating diverse computer vision tasks\n\nAs a result, the SSL algorithms integrated in PixelSSL are compatible with all task codes inherited from the given template. \n\nIn addition, PixelSSL provides the benchmarks for validating semi-supervised learning algorithms for some pixel-level tasks, which now include [semantic segmentation](task/sseg).\n\n\n## News\n- **[Dec 25 2020] PixelSSL v0.1.4 is Released!**  \n  :christmas_tree: ***Merry Christmas!*** :christmas_tree:  \n  v0.1.4 supports the [CutMix](https://arxiv.org/abs/1906.01916) semi-supervised learning algorithm for pixel-wise classification.\n\n- **[Nov 06 2020] PixelSSL v0.1.3 is Released!**  \n  v0.1.3 supports the [CCT](https://arxiv.org/abs/2003.09005) semi-supervised learning algorithm for pixel-wise classification.\n\n- **[Oct 28 2020] PixelSSL v0.1.2 is Released!**  \n  v0.1.2 supports [PSPNet](https://arxiv.org/abs/1612.01105) and its SSL results for semantic segmentation task (check [here](task/sseg)).\n  \n  [[More](docs/updates.md)]\n\n\n## Supported Algorithms and Tasks\nWe are actively updating this project.  \nThe SSL algorithms and demo tasks supported by PixelSSL are summarized in the following table: \n| Algorithms / Tasks | [Segmentation](task/sseg) | Other Tasks | \n| :---: | :---: | :---: |\n| SupOnly | v0.1.0 | Coming Soon |\n| MT [[1]](https://arxiv.org/abs/1703.01780) | v0.1.0 | Coming Soon |\n| AdvSSL [[2]](https://arxiv.org/abs/1802.07934) | v0.1.0 | Coming Soon |\n| S4L [[3]](https://arxiv.org/abs/1905.03670) | v0.1.1 | Coming Soon | \n| CCT [[4]](https://arxiv.org/abs/2003.09005) | v0.1.3 | Coming Soon |\n| GCT [[5]](https://arxiv.org/abs/2008.05258) | v0.1.0 | Coming Soon |\n| CutMix [[6]](https://arxiv.org/abs/1906.01916) | v0.1.4 | Coming Soon |\n\n\n[1] Mean Teachers are Better Role Models: Weight-Averaged Consistency Targets Improve Semi-Supervised Deep Learning Results  \n\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;Antti Tarvainen, and Harri Valpola. NeurIPS 2017.\n\n[2] Adversarial Learning for Semi-Supervised Semantic Segmentation  \n\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;Wei-Chih Hung, Yi-Hsuan Tsai, Yan-Ting Liou, Yen-Yu Lin, and Ming-Hsuan Yang. BMVC 2018.  \n\n[3] S4L: Self-Supervised Semi-Supervised Learning  \n\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;Xiaohua Zhai, Avital Oliver, Alexander Kolesnikov, and Lucas Beyer. ICCV 2019.  \n\n[4] Semi-Supervised Semantic Segmentation with Cross-Consistency Training  \n\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;Yassine Ouali, Céline Hudelot, and Myriam Tami. CVPR 2020.\n\n[5] Guided Collaborative Training for Pixel-wise Semi-Supervised Learning  \n\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;Zhanghan Ke, Di Qiu, Kaican Li, Qiong Yan, and Rynson W.H. Lau. ECCV 2020.\n\n[6] Semi-Supervised Semantic Segmentation Needs Strong, Varied Perturbations  \n\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;Geoff French, Samuli Laine, Timo Aila, Michal Mackiewicz, and Graham Finlayson.  BMVC 2020.\n\n## Installation\nPlease refer to the [Installation](docs/installation.md) document.  \n\n\n## Getting Started\nPlease follow the [Getting Started](docs/getting_started.md) document to run the provided demo tasks.\n\n\n## Tutorials\nWe provide the [API](docs/api.md) document and some tutorials for using PixelSSL.\n- [Tutorial 1 - Implement A New Pixel-wise Semi-Supervised Algorithm](docs/tutorial/tutorial-1.md)\n- [Tutorial 2 - Implement A New Pixel-wise Task Based on the Task Template](docs/tutorial/tutorial-2.md)\n- [Tutorial 3 - Dataset Wrappers for Semi-Supervised Learning](docs/tutorial/tutorial-3.md)\n- [Tutorial 4 - Support More Optimizers and LRSchedulers](docs/tutorial/tutorial-4.md)\n\n\n## License\nThis project is released under the [Apache 2.0 license](LICENSE).\n\n\n## Acknowledgement\nWe thank [City University of Hong Kong](https://www.cityu.edu.hk/) and [SenseTime](https://www.sensetime.com/) for their support to this project.\n\n\n## Citation\nThis project is extended from our ECCV 2020 paper [Guided Collaborative Training for Pixel-wise Semi-Supervised Learning](https://arxiv.org/abs/2008.05258) (GCT). If this codebase or our method helps your research, please cite:\n\n```bibtex\n@InProceedings{ke2020gct,\n  author = {Ke, Zhanghan and Qiu, Di and Li, Kaican and Yan, Qiong and Lau, Rynson W.H.},\n  title = {Guided Collaborative Training for Pixel-wise Semi-Supervised Learning},\n  booktitle = {European Conference on Computer Vision (ECCV)},\n  month = {August},\n  year = {2020},\n}\n```\n\n## Contact\nThis project is currently maintained by Zhanghan Ke ([@ZHKKKe](https://github.com/ZHKKKe)).  \nIf you have any questions, please feel free to contact `kezhanghan@outlook.com`.\n","funding_links":[],"categories":["Papers (with code)","2020 ( Under construction )"],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FZHKKKe%2FPixelSSL","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FZHKKKe%2FPixelSSL","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FZHKKKe%2FPixelSSL/lists"}