{"id":20841187,"url":"https://github.com/sadevans/b2u_sber_implemetation","last_synced_at":"2026-05-09T02:36:33.951Z","repository":{"id":184167179,"uuid":"671409377","full_name":"sadevans/b2u_sber_implemetation","owner":"sadevans","description":"Модификация Blind2Unblind для конкретных данных и задач","archived":false,"fork":false,"pushed_at":"2023-07-27T14:43:23.000Z","size":15989,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-07-31T12:29:23.857Z","etag":null,"topics":["blind2unblind","computer-vision","opencv","pytorch"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","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/sadevans.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,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2023-07-27T08:51:47.000Z","updated_at":"2023-12-26T08:15:48.000Z","dependencies_parsed_at":"2025-07-31T10:31:15.083Z","dependency_job_id":"1038b9ab-0afb-448a-8ff9-745ac3ed17db","html_url":"https://github.com/sadevans/b2u_sber_implemetation","commit_stats":null,"previous_names":["sadevans/b2u_sber_implemetation"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/sadevans/b2u_sber_implemetation","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sadevans%2Fb2u_sber_implemetation","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sadevans%2Fb2u_sber_implemetation/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sadevans%2Fb2u_sber_implemetation/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sadevans%2Fb2u_sber_implemetation/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/sadevans","download_url":"https://codeload.github.com/sadevans/b2u_sber_implemetation/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sadevans%2Fb2u_sber_implemetation/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":32805115,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-08T08:22:46.396Z","status":"online","status_checked_at":"2026-05-09T02:00:06.633Z","response_time":123,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"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":["blind2unblind","computer-vision","opencv","pytorch"],"created_at":"2024-11-18T01:19:07.342Z","updated_at":"2026-05-09T02:36:33.932Z","avatar_url":"https://github.com/sadevans.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# That is implementation of Blind2Unblind: Self-Supervised Image Denoising with Visible Blind Spots for SBER Robotics Lab\n\narticle: [Blind2Unblind](https://arxiv.org/abs/2203.06967)\n\n## Citing Blind2Unblind\n```\n@InProceedings{Wang_2022_CVPR,\n    author    = {Wang, Zejin and Liu, Jiazheng and Li, Guoqing and Han, Hua},\n    title     = {Blind2Unblind: Self-Supervised Image Denoising With Visible Blind Spots},\n    booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},\n    month     = {June},\n    year      = {2022},\n    pages     = {2027-2036}\n}\n```\n\nThe original code is placed here: [github](https://github.com/zejinwang/Blind2Unblind)\n## Installation\nThe model is built in Python3.8.5, PyTorch 1.7.1 in Ubuntu 22.04 environment.\n\n## Data Preparation\n\n### 1. Prepare Training Dataset\n\nPlease put your training dataset under the path: **./b2u_sber_implemetation/data/train**.\n\n\u003c!-- - For processing ImageNet Validation, please run the command\n\n  ```shell\n  python ./dataset_tool.py\n  ```\n\n- For processing SIDD Medium Dataset in raw-RGB, please run the command\n\n  ```shell\n  python ./dataset_tool_raw.py\n  ``` --\u003e\n\n### 2. Prepare Validation Dataset\n\n​\tPlease put your validation dataset under the path: **./b2u_sber_implemetation/data/test**.\n\n## Pretrained Models\nYou can find pre-trained models here: **./b2u_sber_implemetation/pretrained_models**\n\nModels were trained on datasets G-209, Crystal_focus_0_dose_180, G-146\n\n```yaml\n# # For more noisy datasets processing use model firstly trained on G-209\n./pretrained_models/b2u_first.pth\n# Than use model secondly trained on G-209 denoised by first model\n./pretrained_models/b2u_second.pth\n\n\n# # For less noisy images use model trained on Crystal_focus_0_dose_180\n./pretrained_models/b2u_crystal_first.pth\n```\n\n## Train\n* For training your own model please use [SBER_train](https://github.com/sadevans/b2u_sber_implemetation/blob/f2865e86ba95634329dfbdb229182295d3da0425/SBER_train.ipynb#L10)\n\n\n## Test\n\nPlease put your test data in the folder: **./b2u_sber_implemetation/test**\n\n* To test model on images maximum size **768x1024** use [SBER_test_small_images](https://github.com/sadevans/b2u_sber_implemetation/blob/main/SBER_test_small_images.ipynb)\n\n* To test model on large resolution images use [SBER_test_large_images](https://github.com/sadevans/b2u_sber_implemetation/blob/main/SBER_test_large_images.ipynb)\n\nIn this jupyter notebook you can set: \n- your image proportions, \n- crop propotions\n- margin value for cropping and concating without visible joints\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsadevans%2Fb2u_sber_implemetation","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsadevans%2Fb2u_sber_implemetation","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsadevans%2Fb2u_sber_implemetation/lists"}