{"id":25999792,"url":"https://github.com/jinyeying/RaindropClarity","last_synced_at":"2025-03-05T18:41:40.851Z","repository":{"id":248955978,"uuid":"825240501","full_name":"jinyeying/RaindropClarity","owner":"jinyeying","description":"[ECCV2024] \"Raindrop Clarity: A Dual-Focused Dataset for Day and Night Raindrop Removal\", https://arxiv.org/abs/2407.16957","archived":false,"fork":false,"pushed_at":"2025-03-03T17:07:55.000Z","size":42670,"stargazers_count":45,"open_issues_count":2,"forks_count":3,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-03-03T18:23:56.575Z","etag":null,"topics":["badweather","dataset","deep-learning","eccv2024","low-level-vision","rain","raindrop","restoration"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/jinyeying.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}},"created_at":"2024-07-07T08:11:22.000Z","updated_at":"2025-03-03T17:07:59.000Z","dependencies_parsed_at":"2024-08-02T04:27:15.732Z","dependency_job_id":"3f5aecb7-b18c-45dc-ab36-11e21d1515bc","html_url":"https://github.com/jinyeying/RaindropClarity","commit_stats":null,"previous_names":["jinyeying/raindropclarity"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jinyeying%2FRaindropClarity","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jinyeying%2FRaindropClarity/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jinyeying%2FRaindropClarity/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jinyeying%2FRaindropClarity/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/jinyeying","download_url":"https://codeload.github.com/jinyeying/RaindropClarity/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":242083055,"owners_count":20069232,"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":["badweather","dataset","deep-learning","eccv2024","low-level-vision","rain","raindrop","restoration"],"created_at":"2025-03-05T18:40:47.404Z","updated_at":"2025-03-05T18:41:40.846Z","avatar_url":"https://github.com/jinyeying.png","language":"Python","funding_links":[],"categories":["Python"],"sub_categories":[],"readme":"# RaindropClarity (ECCV'2024) (CVPR-NTIRE'2025)\n\n## Introduction\n\u003e [Raindrop Clarity: A Dual-Focused Dataset for Day and Night Raindrop Removal](https://arxiv.org/abs/2407.16957)\u003cbr\u003e\n\u003e European Conference on Computer Vision (ECCV'2024) (CVPR-NTIRE'2025)\n\u003e \n[![arXiv](https://img.shields.io/badge/arXiv-Paper-\u003cCOLOR\u003e.svg)](https://arxiv.org/abs/2407.16957)\n[[Poster]](https://github.com/jinyeying/RaindropClarity/blob/main/poster_slides/RaindropClarity_poster.pdf) \n[[Slides]](https://github.com/jinyeying/RaindropClarity/blob/main/poster_slides/RaindropClarity_PPT.pdf) \n[[Video]](https://www.youtube.com/watch?v=LSGvCuT46XU)\n\n# Workshop and Challenges @ CVPR-NTIRE'2025\n🔥 The first challenge on Day and Night Raindrop Removal for Dual-Focused Images\u003cbr\u003e\n[[Competitions]](https://codalab.lisn.upsaclay.fr/competitions/21345) [[ChallengePage]](https://lixinustc.github.io/CVPR-NTIRE2025-RainDrop-Competition.github.io/)\n\n## RaindropClarity Dataset\n |Day_Train   |[Dropbox](https://www.dropbox.com/scl/fi/qes7r934c10qzb21funoj/DayRainDrop_Train.zip?rlkey=bdqa53wgvmhj9x1yf40q0c1p7\u0026st=4taffjkx\u0026dl=0) | [BaiduPan](https://pan.baidu.com/s/1-vwhYA7jEDPAYHlznhcHCA?pwd=j9ay) code:j9ay |[GoogleDrive](https://drive.google.com/file/d/1lHOumI4wDsJfgGnDPOCXM6h_C1F5ZQEA/view?usp=sharing)|\n |:-----------:| :-----------: | :-----------: |:-----------: |\n |Night_Train|[Dropbox](https://www.dropbox.com/scl/fi/cw3ji53qxy18sepuk6wcp/NightRainDrop_Train.zip?rlkey=r2yn224ryek9wxkbchedeg13j\u0026st=jzo93x80\u0026dl=0)| [BaiduPan](https://pan.baidu.com/s/13x6-UzqxaJG7tKv2WMyMuQ?pwd=hmsw) code:hmsw| [GoogleDrive](https://drive.google.com/file/d/1_ruwsBCzbOEkpqcqHIeKoyCg6P2Sxclr/view?usp=sharing)|\n\n## Validation Data (only for Workshop and Challenges)\n|Day + Night |[Dropbox](https://www.dropbox.com/scl/fo/74s4wpvhrx1ag0rvsuq9q/ALyqJEFapXZ7QBcOyV2uCSs?rlkey=hsu5ktfetpjh986rvkmxexule\u0026st=k9m9wgxw\u0026dl=0) |[BaiduPan](https://pan.baidu.com/s/1gwMC3uN6UrJ23xn24NNVdg?pwd=vali) code:vali | [GoogleDrive](https://drive.google.com/drive/folders/1KFInlF3cSZXxUIK4N9YvCC0B4N9lXAhB?usp=sharing)|\n|:-----------:|:-----------:| :-----------: | :-----------: |\n\n# Abstract @ ECCV’2024\n Existing raindrop removal datasets have two shortcomings. First, they consist of images captured by cameras with a focus on the background, leading to the presence of blurry raindrops. To our knowledge, none of these datasets include images where the focus is specifically on raindrops, which results in a blurry background. Second, these datasets predominantly consist of daytime images, thereby lacking nighttime raindrop scenarios. Consequently, algorithms trained on these datasets may struggle to perform effectively in raindrop-focused or nighttime scenarios. The absence of datasets specifically designed for raindrop-focused and nighttime raindrops constrains research in this area. In this paper, we introduce a large-scale, real-world raindrop removal dataset called Raindrop Clarity. Raindrop Clarity comprises 15,186 high-quality pairs/triplets (raindrops, blur, and background) of images with raindrops and the corresponding clear background images. There are 5,442 daytime raindrop images and 9,744 nighttime raindrop images. Specifically, the 5,442 daytime images include 3,606 raindrop- and 1,836 background-focused images. While the 9,744 nighttime images contain 4,838 raindrop- and 4,906 background-focused images. Our dataset will enable the community to explore background-focused and raindrop-focused images, including challenges unique to daytime and nighttime conditions. \n\n\u003cp align=\"left\"\u003e\n  \u003cimg width=950\" src=\"teaser/RaindropClarity.png\"\u003e\n\u003c/p\u003e\n\n## Pre-trained Models: [BaiduPan](https://pan.baidu.com/s/1tzJX--WD7YsYbpc9nGBQ0w?pwd=i3dg) code:i3dg and [Results](https://pan.baidu.com/s/1kVxJK0HgSDe5pglQ2uTj3A?pwd=outp) code:outp\n| Model Name | Model Dropbox | Model BaiduPan | Results Dropbox | Results BaiduPan |\n| :----: | :-----------: | :----------: |:---------------: |  :----------: |\n| Raindrop + Restoration| [Dropbox](https://www.dropbox.com/scl/fo/oy0s69m4jienlvjpu52wi/ACnxDbcyX2K0trBdEJa4DdQ?rlkey=jn0wkbaf8d4xv8rqixhhuhymy\u0026dl=0) | [BaiduPan](https://pan.baidu.com/s/1tzJX--WD7YsYbpc9nGBQ0w?pwd=i3dg) code:i3dg| [Dropbox](https://www.dropbox.com/scl/fo/mw63y7ss04404pb4vnd1r/AOanObkvsJmUFZEGYmud--c?rlkey=p9hjkawn7vn34hahycxjigv6l\u0026st=aak7ifwh\u0026dl=0) | [BaiduPan](https://pan.baidu.com/s/1kVxJK0HgSDe5pglQ2uTj3A?pwd=outp) code:outp|  \n\n1. the checkpoint RainDrop_DiT_ddpm.pth.tar is patch_size=4, image_size:64,\n[transformer2d.py](https://github.com/jinyeying/RaindropClarity/blob/b825ceeb6f2e1df6d33a5b57fe01dece43d3d98a/models/transformer2d.py#L455), follow patch_size=4\n\n2. the checkpoint RainDrop_DiT2_ddpm.pth.tar is patch_size=2, image_size:64, \n[transformer2d.py](https://github.com/jinyeying/RaindropClarity/blob/b825ceeb6f2e1df6d33a5b57fe01dece43d3d98a/models/transformer2d.py#L452), follow patch_size=2\n\n## Evaluation\n```\npython calculate_psnr_ssim_sid.py\n```\nplease change `base_path`, `time_of_day`, `model_name` accordingly.\n\n## Raindrop-focused or Background-focused?\nThe analysis code is available at [analyse/cal_rf_bf.py](https://github.com/jinyeying/RaindropClarity/blob/main/analyse/cal_rf_bf.py)\n```\npython cal_rf_bf.py\n```\n\u003cp align=\"left\"\u003e\n  \u003cimg width=950\" src=\"teaser/example.png\"\u003e\n\u003c/p\u003e\n\n## Test\n```\nbash run_eval_diffusion_day.sh\n```\n```\nbash run_eval_diffusion_night.sh\n```\nInside the script, please change `model_name` accordingly. \n```\nCUDA_VISIBLE_DEVICES=1 python eval_diffusion_day_dit.py --sid \"$sid\"\n```\n```\nCUDA_VISIBLE_DEVICES=1 python eval_diffusion_day_rdiffusion.py --sid \"$sid\"\n```\n```\nCUDA_VISIBLE_DEVICES=2 python eval_diffusion_day_restomer.py --sid \"$sid\"\n```\n```\nCUDA_VISIBLE_DEVICES=1 python eval_diffusion_day_uformer.py --sid \"$sid\"\n```\n```\nCUDA_VISIBLE_DEVICES=2 python eval_diffusion_day_onego.py --sid \"$sid\"\n```\n```\nCUDA_VISIBLE_DEVICES=1 python eval_diffusion_day_idt.py --sid \"$sid\"\n```\n```\nCUDA_VISIBLE_DEVICES=2 python eval_diffusion_day_icra.py --sid \"$sid\"\n```\n```\nCUDA_VISIBLE_DEVICES=1 python eval_diffusion_day_atgan.py --sid \"$sid\"\n```\n\n## Train\n```\nCUDA_VISIBLE_DEVICES=1,2 python train.py --config daytime_64.yml --test_set Raindrop_DiT\n```\nplease change `daytime_64.yml`,`daytime_128.yml`,`daytime_256.yml` according to `model_name` and `image_size`.\n\n## Acknowledgments\nCode is implemented based [WeatherDiffusion](https://github.com/IGITUGraz/WeatherDiffusion), we would like to thank them.\n\n## License\nThe code and models in this repository are licensed under the MIT License for academic and other non-commercial uses.\u003cbr\u003e\nFor commercial use of the code and models, separate commercial licensing is available. Please contact:\n- Yeying Jin (jinyeying@u.nus.edu)\n- Jonathan Tan (jonathan_tano@nus.edu.sg)\n\n### Citation\nIf this work is useful for your research, please cite our paper. \n```BibTeX\n@inproceedings{jin2024raindrop,\n  title={Raindrop Clarity: A Dual-Focused Dataset for Day and Night Raindrop Removal},\n  author={Jin, Yeying and Li, Xin and Wang, Jiadong and Zhang, Yan and Zhang, Malu},\n  booktitle={European Conference on Computer Vision},\n  pages={1--17},\n  year={2024},\n  organization={Springer}\n}\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjinyeying%2FRaindropClarity","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fjinyeying%2FRaindropClarity","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjinyeying%2FRaindropClarity/lists"}