{"id":26647637,"url":"https://github.com/singlezombie/afldm","last_synced_at":"2025-03-24T23:55:59.819Z","repository":{"id":282041550,"uuid":"947289728","full_name":"SingleZombie/AFLDM","owner":"SingleZombie","description":"[CVPR 2025] Alias-free Latent Diffusion Models official implementation","archived":false,"fork":false,"pushed_at":"2025-03-12T13:17:50.000Z","size":3647,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-03-12T14:22:23.615Z","etag":null,"topics":["artificial-intelligence","computer-vision"],"latest_commit_sha":null,"homepage":"https://zhouyifan.net/AF-LDM-Page/","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/SingleZombie.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE.md","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":"2025-03-12T13:05:18.000Z","updated_at":"2025-03-12T14:12:32.000Z","dependencies_parsed_at":"2025-03-12T14:34:20.912Z","dependency_job_id":null,"html_url":"https://github.com/SingleZombie/AFLDM","commit_stats":null,"previous_names":["singlezombie/afldm"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SingleZombie%2FAFLDM","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SingleZombie%2FAFLDM/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SingleZombie%2FAFLDM/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SingleZombie%2FAFLDM/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/SingleZombie","download_url":"https://codeload.github.com/SingleZombie/AFLDM/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":245372341,"owners_count":20604490,"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":["artificial-intelligence","computer-vision"],"created_at":"2025-03-24T23:55:59.178Z","updated_at":"2025-03-24T23:55:59.799Z","avatar_url":"https://github.com/SingleZombie.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"\n\u003ch1 align=\"center\"\u003e\u003cstrong\u003e[CVPR 2025] Alias-free Latent Diffusion Models \u003c/strong\u003e\u003c/h1\u003e\n  \u003cp align=\"center\"\u003e\n            \u003ca href=\"https://zhouyifan.net/about/\"\u003e Yifan Zhou\u003csup\u003e1\u003c/sup\u003e\u003c/a\u003e\n              \u003ca href=\"https://github.com/xizaoqu\"\u003eZeqi Xiao\u003csup\u003e1\u003c/sup\u003e\u003c/a\t\u003e\n              \u003ca href=\"https://williamyang1991.github.io/\"\u003eShuai Yang\u003csup\u003e2\u003c/sup\u003e\u003c/a\u003e\n              \u003ca href=\"https://xingangpan.github.io/\"\u003eXingang Pan\u003csup\u003e1\u003c/sup\u003e\u003c/a\u003e    \u003cbr\u003e\n    \u003csup\u003e1\u003c/sup\u003eS-Lab, Nanyang Technological University, \u003cbr\u003e \u003csup\u003e2\u003c/sup\u003eWangxuan Institute of Computer Technology, Peking University \n\n\n\u003ch3 align=\"center\"\u003e\n\u003ca href=\"http://zhouyifan.net/AF-LDM-Page/\"\u003e  Project Page \u003c/a\u003e |\n\u003ca href=\"https://arxiv.org/abs/2503.09419\"\u003e  Paper \u003c/a\u003e\n\u003c/h3\u003e\n\nOffical PyTorch implementation of Alias-free latent diffusion models.\n\n## Motivation\n\nhttps://github.com/user-attachments/assets/4fcd0c0f-4c0f-48a9-97dc-e5dcab9dd578\n\nWe found the VAE and denoising network in LDM are not equivariant to fractional shifts. We propose an alias-free framework to improve the fractional shift equivariance of LDM. We demonstrate the effectiveness of our method in various applications, including video editing, frame interpolation, super-resolution and normal estimation. \n\n## TODO\n\n- [ ] Chinese/English blog posts\n- [ ] Refine documents\n- [ ] Training scripts\n\n## Update\n\n* \\[03/2025\\]: Repository created.\n\n## Installation\n\n1. Clone the repository. (Don't forget --recursive. Otherwise, please run git submodule update --init --recursive)\n\n```shell\ngit clone git@github.com:SingleZombie/AFLDM.git --recursive\ncd AFLDM\npip install -e .\n```\n\n2. Install PyTorch in your Python environment.\n\n3. Install pip libraries.\n\n```shell\npip install -r requirements.txt\n```\n\n## Inference\n\nAll the detailed commands are shown inside `.sh` files. \n\n### Unconditional Generation Shift\n\n```shell\nbash shift_ldm_ffhq.sh\n```\n\n### Video Editing\n\nDue to the limitation of our computation resource, the finetuned alias-free Stable Diffusion has a poor generation capacity. It can only perform simple editing.\n\n```shell\nbash video_editing.sh\n```\n\n### Image Interpolation\n\n```shell\nbash image_interpolation.sh\n```\n\n### Super-resolution Shift\n\nThis is not a blind SR. The degradation function is fixed.\n\n```shell\nbash shift_ldm_sr.sh\n```\n\n### Normal Esitmation Shift\n\n```shell\nbash shift_normal_estimation.sh\n```\n\n## Citation\n\n```\n@inproceedings{zhou2025afldm,\n      title={Alias-Free Latent Diffusion Models: Improving Fractional Shift Equivariance of Diffusion Latent Space},\n      author={Zhou, Yifan and Xiao, Zeqi and Yang, Shuai and Pan, Xingang },\n      booktitle = {CVPR},\n      year = {2025},\n    }\n```\n\n## Acknowledgements\n\n* [Diffusers](https://github.com/huggingface/diffusers): Our project is bulit on diffusers.\n* [GMFlow](https://github.com/haofeixu/gmflow): Our flow estimator.\n* [StyleGAN3](https://github.com/NVlabs/stylegan3): For sharing alias-free module implementation. \n* [Alias-Free Convnets](https://github.com/hmichaeli/alias_free_convnets): For sharing alias-free module implementation. \n* [I2SB](https://github.com/NVlabs/I2SB): For sharing SR implementation.\n* [StableNormal](https://github.com/Stable-X/StableNormal): For sharing normal estimation dataset.","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsinglezombie%2Fafldm","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsinglezombie%2Fafldm","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsinglezombie%2Fafldm/lists"}