{"id":27618043,"url":"https://github.com/ant-research/edicho","last_synced_at":"2026-02-13T19:31:12.928Z","repository":{"id":272386266,"uuid":"910142776","full_name":"ant-research/edicho","owner":"ant-research","description":"[ICCV 2025] Edicho: Consistent Image Editing in the Wild","archived":false,"fork":false,"pushed_at":"2025-10-22T17:12:19.000Z","size":23505,"stargazers_count":124,"open_issues_count":8,"forks_count":1,"subscribers_count":13,"default_branch":"main","last_synced_at":"2026-02-09T02:23:38.402Z","etag":null,"topics":["diffusion-models","image-editing"],"latest_commit_sha":null,"homepage":"https://ant-research.github.io/edicho/","language":"Python","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/ant-research.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,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2024-12-30T15:45:12.000Z","updated_at":"2026-01-01T20:33:01.000Z","dependencies_parsed_at":null,"dependency_job_id":"f83e9268-a07a-44d0-8e4c-02b5feed48d6","html_url":"https://github.com/ant-research/edicho","commit_stats":null,"previous_names":["ant-research/edicho","ezioby/edicho"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/ant-research/edicho","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ant-research%2Fedicho","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ant-research%2Fedicho/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ant-research%2Fedicho/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ant-research%2Fedicho/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/ant-research","download_url":"https://codeload.github.com/ant-research/edicho/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ant-research%2Fedicho/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":29415533,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-02-13T06:24:03.484Z","status":"ssl_error","status_checked_at":"2026-02-13T06:23:12.830Z","response_time":78,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"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":["diffusion-models","image-editing"],"created_at":"2025-04-23T04:59:32.798Z","updated_at":"2026-02-13T19:31:12.921Z","avatar_url":"https://github.com/ant-research.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Edicho: Consistent Image Editing in the Wild\n\n\u003e **Edicho: Consistent Image Editing in the Wild** \u003cbr\u003e\n\u003e Qingyan Bai, Hao Ouyang, Yinghao Xu, Qiuyu Wang, Ceyuan Yang, Ka Leong Cheng, Yujun Shen, Qifeng Chen \u003cbr\u003e\n\n\u003cdiv align=center\u003e\n\u003cimg src=\"./docs/assets/teaser.jpg\" width=850px\u003e\n\u003c/div\u003e\n\n**Figure:**  Given two images in the wild, Edicho generates consistent editing versions of them in a zero-shot manner. Our approach\nachieves precise consistency for editing parts (left), objects (middle), and the entire images (right) by leveraging explicit correspondence.\n\n**[**[**Paper**](https://arxiv.org/abs/2412.21079)**]**\n**[**[**Project Page**](https://ant-research.github.io/edicho/)**]**\n\nAs a verified need, consistent editing across in-the-wild images remains a technical challenge arising from various unmanageable factors, like object poses, lighting conditions, and photography environments.\n\u003ci\u003eEdicho\u003c/i\u003e, embodying \"echoing the editing effect\", steps in with a training-free solution based on diffusion models, featuring a fundamental design principle of using \u003cb\u003eexplicit\u003c/b\u003e image correspondence to direct editing.\nSpecifically, the key components include an attention manipulation module and a carefully refined classifier-free guidance (CFG) denoising strategy, both of which take into account the pre-estimated correspondence.\nSuch an inference-time algorithm enjoys a plug-and-play nature and is compatible to most diffusion-based editing methods, such as ControlNet and BrushNet.\nExtensive results demonstrate the efficacy of \u003ci\u003eEdicho\u003c/i\u003e in consistent cross-image editing under diverse settings.\n\n\u003cdiv align=center\u003e\n\u003cimg src=\"./docs/assets/pipeline.jpg\" width=900px\u003e\n\u003c/div\u003e\n\n**Figure:**  Method overview. To achieve consistent editing, we propose a training-free and plug-and-play method that injects the pre-computed correspondence into the pre-trained diffusion models and guides the denoising in the two levels of (a) attention features and (b) noisy latents in classifier-free guidance (CFG).\n\n\n## How to Run\n\n### 1) Environment setup\n\n- Main dependencies (pip install -e .) | [BrushNet extras](examples/brushnet/requirements.txt) | [DIFT env](https://github.com/Tsingularity/dift/blob/main/environment.yml)\n\n### 2) Prepare pretrained model weights\n\n- The editing model: [Sanster/brushnet_segmentation_mask](https://huggingface.co/Sanster/brushnet_segmentation_mask/tree/main)\n- The base diffusion model: `realisticVision_diffusers` (for instance, [this model](https://huggingface.co/emilianJR/RealisticVision_V2)). Set `--base_model_path` to the local path of this model checkpoint directory.\n\n### 3) Run inference\n\nOrganize your inputs under one folder (e.g., `./images/toy`) with images and masks:\n- Images named like `input_*.png` or `input_*.jpg`\n- Masks named like `mask_*.jpg`\n\nThen run:\n\n```bash\npython infer.py \\\n  --base_model_path /path/to/checkpoints \\\n  --brushnet_path Sanster/brushnet_segmentation_mask \\\n  --in_dir ./images/toy \\\n  --save_dir_root ./results \\\n  --prompt_main \"Your prompt\" \\\n  --prompt_add \"Prompt suffix\" \\\n  --num_inference_steps 50 \\\n  --seed 42\n```\n\nOutputs will be saved under `./results/\u003cinput_folder_name\u003e/` as a concatenated image.\n\n\n## BibTeX\n\nIf you find our work helpful for your research, please consider to cite:\n```bibtex\n@inproceedings{bai2024edicho,\n  title     = {Edicho: Consistent Image Editing in the Wild},\n  author    = {Bai, Qingyan and Ouyang, Hao and Xu, Yinghao and Wang, Qiuyu and Yang, Ceyuan and Cheng, Ka Leong and Shen, Yujun and Chen, Qifeng},\n  booktitle = {arXiv preprint arXiv:2412.21079},\n  year      = {2024}\n}\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fant-research%2Fedicho","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fant-research%2Fedicho","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fant-research%2Fedicho/lists"}