{"id":56892,"url":"https://github.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods","name":"Awesome-Diffusion-Model-Based-Image-Editing-Methods","description":"Diffusion Model-Based Image Editing: A Survey (TPAMI 2025)","projects_count":126,"last_synced_at":"2026-08-06T23:00:27.025Z","repository":{"id":218911198,"uuid":"745760153","full_name":"SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods","owner":"SiatMMLab","description":"Diffusion Model-Based Image Editing: A Survey (TPAMI 2025)","archived":false,"fork":false,"pushed_at":"2025-07-15T07:59:36.000Z","size":154216,"stargazers_count":712,"open_issues_count":1,"forks_count":43,"subscribers_count":18,"default_branch":"main","last_synced_at":"2026-07-18T16:03:50.215Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"https://ieeexplore.ieee.org/abstract/document/10884879","language":null,"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/SiatMMLab.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":"2024-01-20T04:01:43.000Z","updated_at":"2026-07-13T03:38:58.000Z","dependencies_parsed_at":null,"dependency_job_id":"296e4edd-2779-4226-b0c0-d03886fffb77","html_url":"https://github.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods","commit_stats":null,"previous_names":["siatmmlab/awesome-diffusion-model-based-image-editing-methods"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SiatMMLab%2FAwesome-Diffusion-Model-Based-Image-Editing-Methods","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SiatMMLab%2FAwesome-Diffusion-Model-Based-Image-Editing-Methods/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SiatMMLab%2FAwesome-Diffusion-Model-Based-Image-Editing-Methods/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SiatMMLab%2FAwesome-Diffusion-Model-Based-Image-Editing-Methods/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/SiatMMLab","download_url":"https://codeload.github.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SiatMMLab%2FAwesome-Diffusion-Model-Based-Image-Editing-Methods/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":36353118,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-08-06T04:43:03.162Z","status":"ssl_error","status_checked_at":"2026-08-06T04:43:02.660Z","response_time":54,"last_error":"SSL_read: 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"}},"created_at":"2024-03-14T22:05:36.032Z","updated_at":"2026-08-06T23:00:27.025Z","primary_language":null,"list_of_lists":false,"displayable":true,"categories":["Training and Finetuning Free","Testing-Time Finetuning","Training-Based"],"sub_categories":["Training and Finetuning Free: Multi-Noise Redirection","Training and Finetuning Free: Inversion/Sampling Modification","Testing-Time Finetuning: Denosing Model Finetuning","Testing-Time Finetuning: Guidance with Hypernetworks","Training and Finetuning Free: Mask Guidance","Training-Based: Pseudo-Target Retrieval with Weak Supervision","Training and Finetuning Free: Attention Modification","Testing-Time Finetuning: Latent Variable Optimization","Training and Finetuning Free: Input Text Refinement","Training-Based: Reference and Attribute Guidance via Self-Supervision","Training-Based: Domain-Specific Editing","Testing-Time Finetuning: Hybrid Finetuning","Training-Based: Instructional Editing via Full Supervision","Testing-Time Finetuning: Embeddings Finetuning","Training-Based: Domain-Specific Editing with Weak Supervision","Training-Based: Reference and Attribute Guided Editing","Training-Based: Instructional Editing"],"readme":"[//]: # (# Diffusion Models for Image Editing)\n\n\u003cp align=\"center\"\u003e\n \u003cimg src=\"./media/title.png\" alt=\"image\" style=\"width:1000px;\"\u003e\n\u003c/p\u003e\n\n[![Awesome](media/badge.svg)](https://github.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods) \n[![License: MIT](media/License-MIT-green.svg)](https://opensource.org/licenses/MIT)\n[![Made With Love](media/Made-With-Love-red.svg)](https://github.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods)\n[![arXiv](media/arXiv-Paper-red.svg)](https://arxiv.org/abs/2402.17525) \n[![visitors](https://visitor-badge.laobi.icu/badge?page_id=SiatMMLab.Awesome-Diffusion-Model-Based-Image-Editing-Methods)](https://visitor-badge.laobi.icu/badge?page_id=SiatMMLab.Awesome-Diffusion-Model-Based-Image-Editing-Methods)\n\nThe repository is based on our survey [Diffusion Model-Based Image Editing: A Survey](https://arxiv.org/pdf/2402.17525.pdf) (TPAMI 2025).\n\nYi Huang*, Jiancheng Huang*, Yifan Liu*, Mingfu Yan*, Jiaxi Lv*, Jianzhuang Liu*, Wei Xiong, He Zhang, Liangliang Cao, Shifeng Chen\n\nShenzhen Institute of Advanced Technology (SIAT), Chinese Academy of Sciences (CAS), Adobe Inc, Apple Inc, Southern University of Science and Technology (SUSTech)\n\n## Abstract\nDenoising diffusion models have emerged as a powerful tool for various image generation and editing tasks, facilitating the synthesis of visual content in an unconditional or input-conditional manner. The core idea behind them is learning to reverse the process of gradually adding noise to images, allowing them to generate high-quality samples from a complex distribution. In this survey, we provide an exhaustive overview of existing methods using diffusion models for image editing, covering both theoretical and practical aspects in the ๏ฌeld. \nWe delve into a thorough analysis and categorization of these works from multiple perspectives, including learning strategies, user-input conditions, and the array of specific editing tasks that can be accomplished.\nIn addition, we pay special attention to image inpainting and outpainting, and explore both earlier traditional context-driven and current multimodal conditional methods, offering a comprehensive analysis of their methodologies.\nTo further evaluate the performance of text-guided image editing algorithms, we propose a systematic benchmark, EditEval, featuring an innovative metric, LMM Score.\nFinally, we address current limitations and envision some potential directions for future research.\n\n## ๐Ÿ”– News!!!\n\n๐Ÿ“Œ We are actively tracking the latest research and welcome contributions to our repository and survey paper. If your studies are relevant, please feel free to contact us.\n\n๐Ÿ“ฐ 2025-02-11: ๐Ÿฅณ Congrats, our paper is accepted by TPAMI 2025!! \n\n๐Ÿ“ฐ 2024-10-25: Our benchmark [EditEval_v2](#benchmark-editeval_v2) is now released.\n\n๐Ÿ“ฐ 2024-03-22: The [template](EditEval_v1/Metric/LMM_Score_GPT4V_Prompt_Template.md) of computing LMM Score using GPT-4V, along with a corresponding [leaderboard](#leaderboard) comparing several leading methods, is released.\n\n๐Ÿ“ฐ 2024-03-14: Our benchmark [EditEval_v1](#benchmark-editeval_v1) is now released.\n\n๐Ÿ“ฐ 2024-03-06: We establish a template for paper submissions. This template is accessible by navigating to the `New Issue` button within `Issues` or by clicking [here](https://github.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/issues/new/choose). Once there, please select the `Paper Submission Form` and complete it following the guidelines provided.\n\n๐Ÿ“ฐ 2024-02-28: Our comprehensive survey paper, summarizing related methods published before February 1, 2024, is now available.\n\n\n\n## ๐Ÿ” BibTeX\nIf you find this work helpful in your research, welcome to cite the paper and give a โญ.\n\n```\n@article{huang2025diffusion,\n title={Diffusion Model-Based Image Editing: A Survey},\n author={Huang, Yi and Huang, Jiancheng and Liu, Yifan and Yan, Mingfu and Lv, Jiaxi and Liu, Jianzhuang and Xiong, Wei and Zhang, He and Cao, Liangliang and Chen, Shifeng},\n journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},\n year={2025},\n publisher={IEEE}\n}\n```\n\n\n\n## Table of contents\n- [Papers](#papers)\n - \u003cdetails\u003e\n \u003csummary\u003eTraining-Based\u003c/summary\u003e\n \n - [Training-Based: Domain-Specific Editing with Weak Supervision](#training-based-domain-specific-editing-with-weak-supervision)\n - [Training-Based: Reference and Attribute Guidance via Self-Supervision](#training-based-reference-and-attribute-guidance-via-self-supervision)\n - [Training-Based: Instructional Editing via Full Supervision](#training-based-instructional-editing-via-full-supervision)\n - [Training-Based: Pseudo-Target Retrieval with Weak Supervision](#training-based-pseudo-target-retrieval-with-weak-supervision)\n \u003c/details\u003e\n - \u003cdetails\u003e\n \u003csummary\u003eTesting-Time Finetuning\u003c/summary\u003e\n \n - [Testing-Time Finetuning: Denosing Model Finetuning](#testing-time-finetuning-denosing-model-finetuning)\n - [Testing-Time Finetuning: Embeddings Finetuning](#testing-time-finetuning-embeddings-finetuning)\n - [Testing-Time Finetuning: Guidance with Hypernetworks](#testing-time-finetuning-guidance-with-hypernetworks)\n - [Testing-Time Finetuning: Latent Variable Optimization](#testing-time-finetuning-latent-variable-optimization)\n - [Testing-Time Finetuning: Hybrid Finetuning](#testing-time-finetuning-hybrid-finetuning)\n \u003c/details\u003e\n - \u003cdetails\u003e\n \u003csummary\u003eTraining and Finetuning Free\u003c/summary\u003e\n \n - [Training and Finetuning Free: Input Text Refinement](#training-and-finetuning-free-input-text-refinement)\n - [Training and Finetuning Free: Inversion/Sampling Modification](#training-and-finetuning-free-inversionsampling-modification)\n - [Training and Finetuning Free: Attention Modification](#training-and-finetuning-free-attention-modification)\n - [Training and Finetuning Free: Mask Guidance](#training-and-finetuning-free-mask-guidance)\n - [Training and Finetuning Free: Multi-Noise Redirection](#training-and-finetuning-free-multi-noise-redirection)\n \u003c/details\u003e\n- [Benchmark EditEval_v1](#benchmark-editeval_v1)\n- [Template of Computing LMM Score](EditEval_v1/Metric/LMM_Score_GPT4V_Prompt_Template.md)\n- [Leaderboard](#leaderboard)\n- [Star History](#star-history)\n\n \n# Papers\n## Training-Based\n\n### Training-Based: Domain-Specific Editing\n| Title | Publication | Date |\n|-------------------------------------------------------------------------------------------|--------------------|--------------|\n| [TexFit: Text-Driven Fashion Image Editing with Diffusion Models](https://ojs.aaai.org/index.php/AAAI/issue/view/584) | AAAI 2024 | 2024.03 |\n| [CycleNet: Rethinking Cycle Consistency in Text-Guided Diffusion for Image Manipulation](https://arxiv.org/abs/2310.13165) | NeurIPS 2023 | 2023.10 |\n| [Stylediffusion: Controllable disentangled style transfer via diffusion models](https://openaccess.thecvf.com/content/ICCV2023/papers/Wang_StyleDiffusion_Controllable_Disentangled_Style_Transfer_via_Diffusion_Models_ICCV_2023_paper.pdf) | ICCV 2023 | 2023.08 |\n| [Hierarchical diffusion autoencoders and disentangled image manipulation](https://openaccess.thecvf.com/content/WACV2024/papers/Lu_Hierarchical_Diffusion_Autoencoders_and_Disentangled_Image_Manipulation_WACV_2024_paper.pdf) | WACV 2024 | 2023.04 |\n| [Towards Real-time Text-driven Image Manipulation with Unconditional Diffusion Models](https://arxiv.org/abs/2304.04344) | arXiv 2023 | 2023.04 |\n| [Fine-grained Image Editing by Pixel-wise Guidance Using Diffusion Models](https://arxiv.org/pdf/2212.02024.pdf) | CVPR workshop 2023 | 2022.12 |\n| [Diffstyler: Controllable dual diffusion for text-driven image stylization](https://arxiv.org/pdf/2211.10682.pdf) | TNNLS 2024 | 2022.11 |\n| [Diffusion Models Already Have A Semantic Latent Space](https://arxiv.org/pdf/2210.10960.pdf) | ICLR 2022 | 2022.10 |\n| [Egsde: Unpaired image-to-image translation via energy-guided stochastic differential equations](https://arxiv.org/pdf/2207.06635.pdf) | NeurIPS 2022 | 2022.07 |\n| [Diffusion autoencoders: Toward a meaningful and decodable representation](https://openaccess.thecvf.com/content/CVPR2022/papers/Preechakul_Diffusion_Autoencoders_Toward_a_Meaningful_and_Decodable_Representation_CVPR_2022_paper.pdf) | CVPR 2022 | 2021.11 |\n| [Unit-ddpm: Unpaired image translation with denoising diffusion probabilistic models](https://arxiv.org/pdf/2104.05358.pdf) | arXiv 2021 | 2021.04 |\n| [Diffusionclip: Text-guided diffusion models for robust image manipulation](https://openaccess.thecvf.com/content/CVPR2022/papers/Kim_DiffusionCLIP_Text-Guided_Diffusion_Models_for_Robust_Image_Manipulation_CVPR_2022_paper.pdf) | CVPR 2022 | 2021.01 |\n\n\n### Training-Based: Reference and Attribute Guided Editing\n\n| Title | Publication | Date |\n|-------------------------------------------------------------------------------------------|--------------------|--------------|\n| [MagicEraser: Erasing Any Objects via Semantics-Aware Control](https://arxiv.org/abs/2410.10207) | ECCV 2024 | 2024.10 |\n| [SmartMask: Context Aware High-Fidelity Mask Generation for Fine-grained Object Insertion and Layout Control](https://arxiv.org/pdf/2312.05039.pdf) | CVPR 2024 | 2023.12 |\n| [A Task is Worth One Word: Learning with Task Prompts for High-Quality Versatile Image Inpainting](http://arxiv.org/abs/2312.03594) | arXiv 2023 | 2023.12 |\n| [DreamInpainter: Text-Guided Subject-Driven Image Inpainting with Diffusion Models](http://arxiv.org/abs/2312.03771) | arXiv 2023 | 2023.12 |\n| [Uni-paint: A Unified Framework for Multimodal Image Inpainting with Pretrained Diffusion Model](https://arxiv.org/abs/2310.07222) | ACM MM 2023 | 2023.10 |\n| [Face Aging via Diffusion-based Editing](https://arxiv.org/abs/2309.11321) | BMVC 2023 | 2023.09 |\n| [Anydoor: Zero-shot object-level image customization](https://arxiv.org/abs/2307.09481) | CVPR 2024 | 2023.07 |\n| [Paste, Inpaint and Harmonize via Denoising: Subject-Driven Image Editing with Pre-Trained Diffusion Model](http://arxiv.org/abs/2306.07596) | ICASSP 2024 | 2023.06 |\n| [Text-to-image editing by image information removal](https://arxiv.org/abs/2305.17489) | WACV 2024 | 2023.05 |\n| [Reference-based Image Composition with Sketch via Structure-aware Diffusion Model](https://arxiv.org/abs/2304.09748) | CVPR workshop 2023 | 2023.04 |\n| [PAIR-Diffusion: A Comprehensive Multimodal Object-Level Image Editor](https://arxiv.org/abs/2303.17546) | CVPR 2024 | 2023.03 |\n| [Imagen editor and editbench: Advancing and evaluating text-guided image inpainting](https://arxiv.org/abs/2212.06909) | CVPR 2023 | 2022.12 |\n| [Smartbrush: Text and shape guided object inpainting with diffusion model](https://arxiv.org/abs/2212.05034) | CVPR 2023 | 2022.12 |\n| [ObjectStitch: Object Compositing With Diffusion Model](http://arxiv.org/abs/2212.00932) | CVPR 2023 | 2022.12 |\n| [Paint by example: Exemplar-based image editing with diffusion models](https://openaccess.thecvf.com/content/CVPR2023/html/Yang_Paint_by_Example_Exemplar-Based_Image_Editing_With_Diffusion_Models_CVPR_2023_paper.html) | CVPR 2023 | 2022.11 |\n\n### Training-Based: Instructional Editing\n| Title | Publication | Date |\n|-------------------------------------------------------------------------------------------|--------------------|--------------|\n| [UIP2P: Unsupervised Instruction-based Image Editing via Edit Reversibility Constraint](https://arxiv.org/abs/2412.15216) | ICCV 2025 | 2024.12 |\n| [FreeEdit: Mask-free Reference-based Image Editing with Multi-modal Instruction](https://arxiv.org/abs/2409.18071) | arXiv 2024 | 2024.09 |\n| [EditWorld: Simulating World Dynamics for Instruction-Following Image Editing](https://arxiv.org/abs/2405.14785) | arXiv 2024 | 2024.05 |\n| [InstructGIE: Towards Generalizable Image Editing](https://arxiv.org/abs/2403.05018) | arXiv 2024 | 2024.03 |\n| [SmartEdit: Exploring Complex Instruction-based Image Editing with Multimodal Large Language Models](http://arxiv.org/abs/2312.06739) | CVPR 2024 | 2023.12 |\n| [InstructAny2Pix: Flexible Visual Editing via Multimodal Instruction Following](http://arxiv.org/abs/2312.06738) | arXiv 2023 | 2023.12 |\n| [Focus on Your Instruction: Fine-grained and Multi-instruction Image Editing by Attention Modulation](http://arxiv.org/abs/2312.10113) | CVPR 2024 | 2023.12 |\n| [Emu edit: Precise image editing via recognition and generation tasks](http://arxiv.org/abs/2311.10089) | arXiv 2023 | 2023.11 |\n| [Guiding instruction-based image editing via multimodal large language models](http://arxiv.org/abs/2309.17102) | ICLR 2024 | 2023.09 |\n| [Instructdiffusion: A generalist modeling interface for vision tasks](https://arxiv.org/abs/2309.03895) | CVPR 2024 | 2023.09 |\n| [MoEController: Instruction-based Arbitrary Image Manipulation with Mixture-of-Expert Controllers](https://arxiv.org/abs/2309.04372) | arXiv 2023 | 2023.09 |\n| [ImageBrush: Learning Visual In-Context Instructions for Exemplar-Based Image Manipulation](http://arxiv.org/abs/2308.00906) | NeurIPS 2023 | 2023.08 |\n| [Inst-Inpaint: Instructing to Remove Objects with Diffusion Models](http://arxiv.org/abs/2304.03246) | arXiv 2023 | 2023.04 |\n| [HIVE: Harnessing Human Feedback for Instructional Visual Editing](https://arxiv.org/abs/2303.09618) | CVPR 2024 | 2023.03 |\n| [DialogPaint: A Dialog-based Image Editing Model](http://arxiv.org/abs/2303.10073) | arXiv 2023 | 2023.01 |\n| [Learning to Follow Object-Centric Image Editing Instructions Faithfully](https://aclanthology.org/2023.findings-emnlp.646/) | EMNLP 2023 | 2023.01 |\n| [Instructpix2pix: Learning to follow image editing instructions](https://openaccess.thecvf.com/content/CVPR2023/html/Brooks_InstructPix2Pix_Learning_To_Follow_Image_Editing_Instructions_CVPR_2023_paper.html) | CVPR 2023 | 2022.11 |\n\n### Training-Based: Pseudo-Target Retrieval-Based Editing\n\n| Title | Publication | Date |\n|-------------------------------------------------------------------------------------------|--------------------|--------------|\n| [Text-Driven Image Editing via Learnable Regions](https://arxiv.org/abs/2311.16432) | CVPR 2024 | 2023.11 |\n| [iEdit: Localised Text-guided Image Editing with Weak Supervision](https://arxiv.org/abs/2305.05947) | arXiv 2023 | 2023.05 |\n| [ChatFace: Chat-Guided Real Face Editing via Diffusion Latent Space Manipulation](https://arxiv.org/abs/2305.14742) | arXiv 2023 | 2023.05 |\n\n\n## Testing-Time Finetuning\n\n### Testing-Time Finetuning: Denosing Model Finetuning\n\n| Title | Publication | Date |\n|-------------------------------------------------------------------------------------------|--------------------|--------------|\n| [Kv inversion: Kv embeddings learning for text-conditioned real image action editing](https://link.springer.com/chapter/10.1007/978-981-99-8429-9_14) | arXiv 2023 | 2023.09 |\n| [Custom-edit: Text-guided image editing with customized diffusion models](https://arxiv.org/abs/2305.15779) | CVPR workshop 2023 | 2023.05 |\n| [Unitune: Text-driven image editing by fine tuning an image generation model on a single image](https://arxiv.org/abs/2210.09477) | ACM TOG 2023 | 2022.10 |\n\n\n### Testing-Time Finetuning: Embeddings Finetuning\n\n| Title | Publication | Date |\n|-------------------------------------------------------------------------------------------|--------------------|--------------|\n| [Dynamic Prompt Learning: Addressing Cross-Attention Leakage for Text-Based Image Editing](https://openreview.net/forum?id=5UXXhVI08r) | NeurIPS 2023 | 2023.09 |\n| [Prompt Tuning Inversion for Text-Driven Image Editing Using Diffusion Models](http://arxiv.org/abs/2305.04441) | ICCV 2023 | 2023.05 |\n| [Uncovering the Disentanglement Capability in Text-to-Image Diffusion Models](https://openaccess.thecvf.com/content/CVPR2023/html/Wu_Uncovering_the_Disentanglement_Capability_in_Text-to-Image_Diffusion_Models_CVPR_2023_paper.html) | CVPR 2023 | 2022.12 |\n| [Null-text inversion for editing real images using guided diffusion models](https://openaccess.thecvf.com/content/CVPR2023/html/Mokady_NULL-Text_Inversion_for_Editing_Real_Images_Using_Guided_Diffusion_Models_CVPR_2023_paper.html) | CVPR 2023 | 2022.11 |\n\n\n\n### Testing-Time Finetuning: Guidance with Hypernetworks\n\n| Title | Publication | Date |\n|-------------------------------------------------------------------------------------------|--------------------|--------------|\n| [StyleDiffusion: Prompt-Embedding Inversion for Text-Based Editing](https://arxiv.org/abs/2303.15649) | arXiv 2023 | 2023.05 |\n| [Inversion-based creativity transfer with diffusion models](https://arxiv.org/abs/2211.13203) | CVPR 2023 | 2022.11 |\n\n\n\n### Testing-Time Finetuning: Latent Variable Optimization\n\n| Title | Publication | Date |\n|-------------------------------------------------------------------------------------------|--------------------|--------------|\n| [StableDrag: Stable Dragging for Point-based Image Editing](https://arxiv.org/abs/2403.04437) | arXiv 2024 | 2024.03 |\n| [FreeDrag: Feature Dragging for Reliable Point-based Image Editing](https://arxiv.org/abs/2307.04684) | CVPR 2024 | 2023.12 |\n| [Contrastive Denoising Score for Text-guided Latent Diffusion Image Editing](https://arxiv.org/abs/2311.18608) | CVPR 2024 | 2023.11 |\n| [MagicRemover: Tuning-free Text-guided Image inpainting with Diffusion Models](https://arxiv.org/abs/2310.02848) | arXiv 2023 | 2023.10 |\n| [Dragondiffusion: Enabling drag-style manipulation on diffusion models](https://arxiv.org/abs/2307.02421) | ICLR 2024 | 2023.07 |\n| [DragDiffusion: Harnessing Diffusion Models for Interactive Point-based Image Editing](https://arxiv.org/abs/2306.14435) | CVPR 2024 | 2023.06 |\n| [Delta denoising score](https://openaccess.thecvf.com/content/ICCV2023/html/Hertz_Delta_Denoising_Score_ICCV_2023_paper.html) | ICCV 2023 | 2023.04 |\n| [Directed Diffusion: Direct Control of Object Placement through Attention Guidance](https://arxiv.org/pdf/2302.13153.pdf) | AAAI 2024 | 2023.02 |\n| [Diffusion-based Image Translation using disentangled style and content representation](https://arxiv.org/abs/2209.15264) | ICLR 2022 | 2022.09 |\n\n\n### Testing-Time Finetuning: Hybrid Finetuning\n\n| Title | Publication | Date |\n|-------------------------------------------------------------------------------------------|--------------------|--------------|\n| [Forgedit: Text Guided Image Editing via Learning and Forgetting](https://arxiv.org/abs/2309.10556) | arXiv 2023 | 2023.09 |\n| [LayerDiffusion: Layered Controlled Image Editing with Diffusion Models](https://arxiv.org/abs/2305.18676) | arXiv 2023 | 2023.05 |\n| [Sine: Single image editing with text-to-image diffusion models](https://openaccess.thecvf.com/content/CVPR2023/html/Zhang_SINE_SINgle_Image_Editing_With_Text-to-Image_Diffusion_Models_CVPR_2023_paper.html) | CVPR 2023 | 2022.12 |\n| [Imagic: Text-Based Real Image Editing With Diffusion Models](https://openaccess.thecvf.com/content/CVPR2023/html/Kawar_Imagic_Text-Based_Real_Image_Editing_With_Diffusion_Models_CVPR_2023_paper.html) | CVPR 2023 | 2022.10 |\n\n\n\n## Training and Finetuning Free\n\n### Training and Finetuning Free: Input Text Refinement\n\n| Title | Publication | Date |\n|-------------------------------------------------------------------------------------------|--------------------|--------------|\n| [User-friendly Image Editing with Minimal Text Input: Leveraging Captioning and Injection Techniques](https://arxiv.org/abs/2306.02717) | arXiv 2023 | 2023.06 |\n| [ReGeneration Learning of Diffusion Models with Rich Prompts for Zero-Shot Image Translation](https://arxiv.org/abs/2305.04651) | arXiv 2023 | 2023.05 |\n| [InstructEdit: Improving Automatic Masks for Diffusion-based Image Editing With User Instructions](https://arxiv.org/abs/2305.18047) | arXiv 2023 | 2023.05 |\n| [Preditor: Text guided image editing with diffusion prior](https://arxiv.org/abs/2302.07979) | arXiv 2023 | 2023.02 |\n\n### Training and Finetuning Free: Inversion/Sampling Modification\n\n| Title | Publication | Date |\n|-------------------------------------------------------------------------------------------|--------------------|--------------|\n| [FireFlow: Fast Inversion of Rectified Flow for Image Semantic Editing](https://arxiv.org/abs/2412.07517) | arXiv 2024 | 2024.12 |\n| [Inversion-Free Image Editing with Natural Language](https://arxiv.org/abs/2312.04965) | CVPR 2024 | 2023.12 |\n| [Fixed-point Inversion for Text-to-image diffusion models](https://arxiv.org/abs/2312.12540) | arXiv 2023 | 2023.12 |\n| [Tuning-Free Inversion-Enhanced Control for Consistent Image Editing](https://arxiv.org/abs/2312.14611) | arXiv 2023 | 2023.12 |\n| [The Blessing of Randomness: SDE Beats ODE in General Diffusion-based Image Editing](https://arxiv.org/abs/2311.01410) | ICLR 2024 | 2023.11 |\n| [LEDITS++: Limitless Image Editing using Text-to-Image Models](https://arxiv.org/abs/2311.16711) | CVPR 2024 | 2023.11 |\n| [A latent space of stochastic diffusion models for zero-shot image editing and guidance](https://openaccess.thecvf.com/content/ICCV2023/html/Wu_A_Latent_Space_of_Stochastic_Diffusion_Models_for_Zero-Shot_Image_ICCV_2023_paper.html) | ICCV 2023 | 2023.10 |\n| [Effective real image editing with accelerated iterative diffusion inversion](https://openaccess.thecvf.com/content/ICCV2023/html/Pan_Effective_Real_Image_Editing_with_Accelerated_Iterative_Diffusion_Inversion_ICCV_2023_paper.html) | ICCV 2023 | 2023.09 |\n| [Fec: Three finetuning-free methods to enhance consistency for real image editing](https://arxiv.org/abs/2309.14934) | arXiv 2023 | 2023.09 |\n| [Iterative multi-granular image editing using diffusion models](https://arxiv.org/abs/2309.00613) | WACV 2024 | 2023.09 |\n| [ProxEdit: Improving Tuning-Free Real Image Editing With Proximal Guidance](https://arxiv.org/abs/2306.05414) | WACV 2024 | 2023.06 |\n| [Diffusion self-guidance for controllable image generation](https://arxiv.org/abs/2306.00986) | NeurIPS 2023 | 2023.06 |\n| [Diffusion Brush: A Latent Diffusion Model-based Editing Tool for AI-generated Images](https://arxiv.org/abs/2306.00219) | arXiv 2023 | 2023.06 |\n| [Null-text guidance in diffusion models is secretly a cartoon-style creator](https://arxiv.org/abs/2305.06710) | ACM MM 2023 | 2023.05 |\n| [Negative-prompt Inversion: Fast Image Inversion for Editing with Text-guided Diffusion Models](https://arxiv.org/abs/2305.16807) | arXiv 2023 | 2023.05 |\n| [An Edit Friendly DDPM Noise Space: Inversion and Manipulations](https://github.com/inbarhub/DDPM_inversion) | CVPR 2024 | 2023.04 |\n| [Training-Free Content Injection Using H-Space in Diffusion Models](https://arxiv.org/abs/2303.15403) | WACV 2024 | 2023.03 |\n| [Edict: Exact diffusion inversion via coupled transformations](https://openaccess.thecvf.com/content/CVPR2023/html/Wallace_EDICT_Exact_Diffusion_Inversion_via_Coupled_Transformations_CVPR_2023_paper.html) | CVPR 2023 | 2022.11 |\n| [Direct inversion: Optimization-free text-driven real image editing with diffusion models](https://arxiv.org/abs/2211.07825) | arXiv 2022 | 2022.11 |\n\n\n### Training and Finetuning Free: Attention Modification\n| Title | Publication | Date |\n|-------------------------------------------------------------------------------------------|--------------------|--------------|\n| [KV-Edit: Training-Free Image Editing for Precise Background Preservation](https://arxiv.org/abs/2502.17363) | arXiv 2025 | 2025.02 |\n| [LIME: Localized Image Editing via Attention Regularization in Diffusion Models](https://arxiv.org/abs/2312.09256) | WACV 2025 | 2024.12 |\n| [Towards Understanding Cross and Self-Attention in Stable Diffusion for Text-Guided Image Editing](https://arxiv.org/abs/2403.03431) | CVPR 2024 | 2024.03 |\n| [HD-Painter: High-Resolution and Prompt-Faithful Text-Guided Image Inpainting with Diffusion Models](https://arxiv.org/abs/2312.14091) | arXiv 2023 | 2023.12 |\n| [Tf-icon: Diffusion-based training-free cross-domain image composition](https://arxiv.org/abs/2307.12493) | ICCV 2023 | 2023.07 |\n| [Energy-Based Cross Attention for Bayesian Context Update in Text-to-Image Diffusion Models](https://arxiv.org/abs/2306.09869) | NeurIPS 2023 | 2023.06 |\n| [Conditional Score Guidance for Text-Driven Image-to-Image Translation](https://arxiv.org/abs/2305.18007) | NeurIPS 2023 | 2023.05 |\n| [MasaCtrl: Tuning-Free Mutual Self-Attention Control for Consistent Image Synthesis and Editing](https://arxiv.org/abs/2304.08465) | ICCV 2023 | 2023.04 |\n| [Localizing Object-level Shape Variations with Text-to-Image Diffusion Models](https://arxiv.org/abs/2303.11306) | ICCV 2023 | 2023.03 |\n| [Zero-shot image-to-image translation](https://dl.acm.org/doi/abs/10.1145/3588432.3591513) | ACM SIGGRAPH 2023 | 2023.02 |\n| [Shape-Guided Diffusion With Inside-Outside Attention](https://openaccess.thecvf.com/content/WACV2024/html/Park_Shape-Guided_Diffusion_With_Inside-Outside_Attention_WACV_2024_paper.html) | WACV 2024 | 2022.12 |\n| [Plug-and-play diffusion features for text-driven image-to-image translation](https://openaccess.thecvf.com/content/CVPR2023/html/Tumanyan_Plug-and-Play_Diffusion_Features_for_Text-Driven_Image-to-Image_Translation_CVPR_2023_paper.html) | CVPR 2023 | 2022.11 |\n| [Prompt-to-prompt image editing with cross attention control](https://openreview.net/forum?id=_CDixzkzeyb) | ICLR 2023 | 2022.08 |\n\n\n### Training and Finetuning Free: Mask Guidance\n\n| Title | Publication | Date |\n|-------------------------------------------------------------------------------------------|---------------|---------|\n| [Grounded-Instruct-Pix2Pix: Improving Instruction Based Image Editing with Automatic Target Grounding](https://ieeexplore.ieee.org/document/10446377) | ICASSP 2024 | 2024.03 |\n| [MAG-Edit: Localized Image Editing in Complex Scenarios via Mask-Based Attention-Adjusted Guidance](https://arxiv.org/abs/2312.11396) | ACM MM 2024 | 2023.12 |\n| [ZONE: Zero-Shot Instruction-Guided Local Editing](https://arxiv.org/abs/2312.16794) | CVPR 2024 | 2023.12 |\n| [Watch your steps: Local image and scene editing by text instructions](https://arxiv.org/abs/2308.08947) | arXiv 2023 | 2023.08 |\n| [Energy-Based Cross Attention for Bayesian Context Update in Text-to-Image Diffusion Models](https://arxiv.org/abs/2306.09869) | NeurIPS 2023 | 2023.06 |\n| [Differential Diffusion: Giving Each Pixel Its Strength](https://arxiv.org/abs/2306.00950) | arXiv 2023 | 2023.06 |\n| [PFB-Diff: Progressive Feature Blending Diffusion for Text-driven Image Editing](https://arxiv.org/abs/2306.16894) | arXiv 2023 | 2023.06 |\n| [FISEdit: Accelerating Text-to-image Editing via Cache-enabled Sparse Diffusion Inference](https://arxiv.org/abs/2305.17423) | AAAI 2024 | 2023.05 |\n| [Inpaint anything: Segment anything meets image inpainting](https://arxiv.org/abs/2304.06790) | arXiv 2023 | 2023.04 |\n| [Region-aware diffusion for zero-shot text-driven image editing](https://arxiv.org/abs/2302.11797) | CVM 2023 | 2023.02 |\n| [Text-guided mask-free local image retouching](https://ieeexplore.ieee.org/abstract/document/10219704) | ICME 2023 | 2022.12 |\n| [Blended diffusion for text-driven editing of natural images](https://openaccess.thecvf.com/content/CVPR2022/papers/Avrahami_Blended_Diffusion_for_Text-Driven_Editing_of_Natural_Images_CVPR_2022_paper.pdf) | CVPR 2022 | 2021.11 |\n| [DiffEdit: Diffusion-based semantic image editing with mask guidance](https://openreview.net/forum?id=3lge0p5o-M-) | ICLR 2023 | 2022.10 |\n| [Blended latent diffusion](https://arxiv.org/abs/2206.02779) | SIGGRAPH 2023 | 2022.06 |\n\n### Training and Finetuning Free: Multi-Noise Redirection\n\n\n| Title | Publication | Date |\n|-------------------------------------------------------------------------------------------|--------------------|--------------|\n| [Object-aware Inversion and Reassembly for Image Editing](https://arxiv.org/abs/2310.12149) | ICLR 2024 | 2023.10 |\n| [Ledits: Real image editing with ddpm inversion and semantic guidance](https://arxiv.org/abs/2307.00522) | arXiv 2023 | 2023.07 |\n| [Sega: Instructing diffusion using semantic dimensions](https://arxiv.org/abs/2301.12247) | NeurIPS 2023 | 2023.01 |\n| [The stable artist: Steering semantics in diffusion latent space](https://arxiv.org/abs/2212.06013) | arXiv 2022 | 2022.12 |\n\n# Benchmark EditEval_v1\n\n\n**EditEval_v1** is a benchmark tailored for evaluation of general diffusion-model based image editing algorithms. It contains 50 high-quality images selected from [Unsplash](https://unsplash.com/), each accompanied by a source text prompt, a target editing prompt, and a text editing instruction generated by GPT-4V. This benchmark covers seven most popular specific editing tasks across semantic, stylistic and structural editing defined in our paper: *object addition*, *object replacement*, *object removal*, *background change*, *overall style change*, *texture change*, and *action change*. Click [here](EditEval_v1/Dataset) to download this dataset!\n\n# Benchmark EditEval_v2\n\n**EditEval_v2** is an enhanced benchmark designed to evaluate general diffusion-model-based image editing algorithms. This version expands upon its predecessor by including 150 high-quality images selected from [Unsplash](https://unsplash.com/). Each image is paired with a source text prompt, a target editing prompt, and a text editing instruction generated by GPT-4V. EditEval_v2 continues to cover the seven most popular specific editing tasks across semantic, stylistic, and structural editing as defined in our paper: *object addition*, *object replacement*, *object removal*, *background change*, *overall style change*, *texture change*, and *action change*. Click [here](https://drive.google.com/file/d/13PCkf_NmkoMsO4E-05eiv714ykd_69QA/view?usp=sharing) to download this dataset!\n\n# Leaderboard\n\nTo facilitate a user-friendly application of LMM Score, [here](EditEval_v1/Metric/LMM_Score_GPT4V_Prompt_Template.md) we provide a comprehensive template for its implementation in GPT-4V. This template comes with step-by-step instructions and all required materials, making it easy for users to apply. Additionally, we construct a leaderboard comparing various representative methods evaluated using LMM Score on our [EditEval_v1](EditEval_v1/Dataset) benchmark, which can be found [here](Leaderboard.md).\n\n# Star History\n\n![Star History Chart](https://api.star-history.com/svg?repos=SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods\u0026type=Date)\n","projects_url":"https://awesome.ecosyste.ms/api/v1/lists/siatmmlab%2Fawesome-diffusion-model-based-image-editing-methods/projects"}