{"id":19832105,"url":"https://github.com/iceclear/clip-iqa","last_synced_at":"2025-04-06T15:12:32.511Z","repository":{"id":63708752,"uuid":"513791598","full_name":"IceClear/CLIP-IQA","owner":"IceClear","description":"[AAAI 2023] Exploring CLIP for Assessing the Look and Feel of Images","archived":false,"fork":false,"pushed_at":"2023-10-27T18:37:30.000Z","size":31731,"stargazers_count":390,"open_issues_count":24,"forks_count":20,"subscribers_count":4,"default_branch":"v2-3.8","last_synced_at":"2025-03-30T13:08:24.445Z","etag":null,"topics":["clip","iqa"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/IceClear.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":"2022-07-14T06:50:27.000Z","updated_at":"2025-03-30T07:30:36.000Z","dependencies_parsed_at":"2023-01-23T18:31:15.183Z","dependency_job_id":"7c1243f3-d9d1-4ca9-baf9-176294fddf8e","html_url":"https://github.com/IceClear/CLIP-IQA","commit_stats":null,"previous_names":[],"tags_count":1,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/IceClear%2FCLIP-IQA","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/IceClear%2FCLIP-IQA/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/IceClear%2FCLIP-IQA/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/IceClear%2FCLIP-IQA/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/IceClear","download_url":"https://codeload.github.com/IceClear/CLIP-IQA/tar.gz/refs/heads/v2-3.8","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247500469,"owners_count":20948880,"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":["clip","iqa"],"created_at":"2024-11-12T11:36:30.362Z","updated_at":"2025-04-06T15:12:32.458Z","avatar_url":"https://github.com/IceClear.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"\n## Exploring CLIP for Assessing the Look and Feel of Images (AAAI 2023)\n\n[Paper](https://arxiv.org/abs/2207.12396)\n\n\n![visitors](https://visitor-badge.laobi.icu/badge?page_id=IceClear/CLIP-IQA)\n\n\n[Jianyi Wang](https://iceclear.github.io/), [Kelvin C.K. Chan](https://ckkelvinchan.github.io/), [Chen Change Loy](https://www.mmlab-ntu.com/person/ccloy/)\n\nS-Lab, Nanyang Technological University\n\n\u003cimg src=\"https://user-images.githubusercontent.com/22350795/202890659-2b73008f-fc0d-49c6-8bc8-0c1f07df5d36.png\" width=\"800px\"/\u003e\n\n### TODO\n- [ ] Colab demo\n- [x] ~~MMEditing update~~\n- [x] ~~Code release~~\n\n### Dependencies and Installation\nThe same as [MMEditing](https://mmediting.readthedocs.io/en/latest/install.html), support the latest version 0.16.1.\n```\n# Create a conda environment and activate it\nconda create -n clipiqa python=3.8 -y\nconda activate clipiqa\n# Install PyTorch following official instructions, e.g.\nconda install pytorch=1.10 torchvision cudatoolkit=11.3 -c pytorch\n# Install pre-built MMCV using MIM.\npip3 install openmim\nmim install mmcv-full==1.5.0\n# Install CLIP-IQA from the source code.\ngit clone git@github.com:IceClear/CLIP-IQA.git\ncd CLIP-IQA\npip install -r requirements.txt\npip install -e .\n```\n\n### Running Examples\n\n#### Test CLIP-IQA on [KonIQ-10k](http://database.mmsp-kn.de/koniq-10k-database.html)\n\n```\npython demo/clipiqa_koniq_demo.py\n```\n\n#### Test CLIP-IQA on [Live-iWT](https://live.ece.utexas.edu/research/ChallengeDB/index.html)\n\n```\npython demo/clipiqa_liveiwt_demo.py\n```\n\n#### Train CLIP-IQA+ on KonIQ-10k\n\n```\n# Support dist training as MMEditing\npython tools/train.py configs/clipiqa/clipiqa_coop_koniq.py\n```\n\n#### Test CLIP-IQA+ on KonIQ-10k ([Checkpoint](https://github.com/IceClear/CLIP-IQA/releases/download/Pretrained/iter_80000.pth))\n\n```\npython demo/clipiqa_koniq_demo.py --config configs/clipiqa/clipiqa_coop_koniq.py --checkpoint ./iter_80000.pth\n```\n\n[Note] You may change prompts for different datasets, please refer to [config files](https://github.com/IceClear/CLIP-IQA/blob/main/configs/clipiqa/clipiqa_attribute_test.py#L11) for details.\n\n[Note] For testing on a single image, please refer to [here](https://github.com/IceClear/CLIP-IQA/tree/main/demo/clipiqa_single_image_demo.py) for details.\n\n### Other Implementations\n- [torchmetrics](https://lightning.ai/docs/torchmetrics/stable/multimodal/clip_iqa.html)\n- [IQA-Pytorch](https://github.com/chaofengc/IQA-PyTorch)\n\n### Demo\n\n#### :sparkles: Versatile Quality Assessment\n\u003cimg src=\"https://user-images.githubusercontent.com/22350795/202886677-63c6af8d-4ae8-4c88-a6e6-b0b980738634.png\" width=\"800px\"/\u003e\n\n#### :sparkles: Demo for IQA on SPAQ\n\u003cimg src=\"assets/SPAQ-exp.png\" width=\"800px\"/\u003e\n\n#### :sparkles: Demo for Abstract Perception on AVA\n\u003cimg src=\"assets/AVA-exp.png\" width=\"800px\"/\u003e\n\nFor more evaluation, please refer to our [paper](https://arxiv.org/abs/2207.12396) for details.\n\n### Citation\nIf our work is useful for your research, please consider citing:\n\n    @inproceedings{wang2022exploring,\n        author = {Wang, Jianyi and Chan, Kelvin CK and Loy, Chen Change},\n        title = {Exploring CLIP for Assessing the Look and Feel of Images},\n        booktitle = {AAAI},\n        year = {2023}\n    }\n\n### License\n\nThis project is licensed under \u003ca rel=\"license\" href=\"https://github.com/IceClear/CLIP-IQA/blob/master/LICENSE\"\u003eNTU S-Lab License 1.0\u003c/a\u003e. Redistribution and use should follow this license.\n\n### Acknowledgement\n\nThis project is based on [MMEditing](https://github.com/open-mmlab/mmediting) and [CLIP](https://github.com/openai/CLIP). Thanks for their awesome works.\n\n### Contact\nIf you have any question, please feel free to reach me out at `iceclearwjy@gmail.com`.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ficeclear%2Fclip-iqa","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ficeclear%2Fclip-iqa","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ficeclear%2Fclip-iqa/lists"}