├── .github ├── ISSUE_TEMPLATE │ └── paper-submission-form.yaml └── workflows │ └── main.yml ├── EditEval_v1 ├── Dataset │ ├── action_change │ │ ├── 1.jpeg │ │ ├── 2.jpg │ │ ├── 3.jpeg │ │ └── 4.jpg │ ├── background_change │ │ ├── 1.jpg │ │ ├── 2.jpg │ │ ├── 3.jpg │ │ ├── 4.jpg │ │ ├── 5.jpg │ │ ├── 6.jpg │ │ ├── 7.jpg │ │ └── 8.jpg │ ├── editing_prompts_collection.xlsx │ ├── object_addition │ │ ├── 1.jpg │ │ ├── 2.jpeg │ │ ├── 3.jpg │ │ ├── 4.jpg │ │ ├── 5.jpg │ │ ├── 6.jpg │ │ ├── 7.jpg │ │ └── 8.jpg │ ├── object_removel │ │ ├── 1.jpg │ │ ├── 2.jpg │ │ ├── 3.jpg │ │ ├── 4.jpeg │ │ ├── 5.jpeg │ │ ├── 6.jpg │ │ ├── 7.jpg │ │ └── 8.jpg │ ├── object_replacement │ │ ├── 1.jpg │ │ ├── 2.jpg │ │ ├── 3.jpg │ │ ├── 4.jpg │ │ ├── 5.jpg │ │ ├── 6.jpg │ │ ├── 7.jpg │ │ └── 8.jpg │ ├── style_change │ │ ├── 1.jpg │ │ ├── 2.jpg │ │ ├── 3.jpg │ │ ├── 4.jpg │ │ ├── 5.jpg │ │ ├── 6.jpg │ │ ├── 7.jpg │ │ └── 8.jpg │ └── texture_change │ │ ├── 1.jpg │ │ ├── 2.jpg │ │ ├── 3.jpg │ │ ├── 4.jpg │ │ ├── 5.jpg │ │ └── 6.jpg └── Metric │ └── LMM_Score_GPT4V_Prompt_Template.md ├── LICENSE ├── Leaderboard.md ├── README.MD └── media ├── License-MIT-green.svg ├── Made-With-Love-red.svg ├── arXiv-Paper-red.svg ├── arXiv.svg ├── badge.svg ├── bg_change ├── 6_AIDI.png ├── 6_DDS.png ├── 6_InstructionPix2Pix.png ├── 6_LEDITS_PLUSPLUS.png ├── 6_SINE.png ├── 6_edict.png ├── 6_null_text.png ├── 6_proxedit.png └── 6_src.jpg └── title.png /.github/ISSUE_TEMPLATE/paper-submission-form.yaml: -------------------------------------------------------------------------------- 1 | name: Paper Submission Form 2 | description: Use this form to submit a paper for this repository. 3 | title: "Paper Submission " 4 | labels: ["paper-submission"] 5 | assignees: ['huangjch526'] 6 | body: 7 | - type: input 8 | id: paper-title 9 | attributes: 10 | label: Paper Title 11 | description: Enter the title of your paper. 12 | placeholder: "Your paper title here" 13 | validations: 14 | required: true 15 | 16 | - type: input 17 | id: publication-venue 18 | attributes: 19 | label: Publication 20 | description: "Please specify where your paper is published." 21 | placeholder: "arXiv 2024, etc." 22 | validations: 23 | required: true 24 | 25 | - type: input 26 | id: paper-link 27 | attributes: 28 | label: Paper Link 29 | description: Provide an arXiv link or publication link for your paper. 30 | placeholder: "http(s)://" 31 | validations: 32 | required: true 33 | 34 | - type: dropdown 35 | id: main-category 36 | attributes: 37 | label: Main Method Category 38 | description: Select the main method category of your paper. 39 | options: 40 | - Training-Based 41 | - Testing-Time Finetuning 42 | - Training and Finetuning Free 43 | validations: 44 | required: true 45 | 46 | - type: dropdown 47 | id: training-based-specifics 48 | attributes: 49 | label: Training-Based Specifics (Optional) 50 | description: "If you select 'Training-Based', please select a more specific category. If you prefer not to specify, it's okay; we will classify it for you." 51 | options: 52 | - Domain-Specific Editing with Weak Supervision 53 | - Reference and Attribute Guidance via Self-Supervision 54 | - Instructional Editing via Full Supervision 55 | - Pseudo-Target Retrieval with Weak Supervision 56 | validations: 57 | required: false 58 | 59 | - type: dropdown 60 | id: testing-time-finetuning-specifics 61 | attributes: 62 | label: Testing-Time Finetuning Specifics (Optional) 63 | description: "If you select 'Testing-Time Finetuning', please select a more specific category. If you prefer not to specify, it's okay; we will classify it for you." 64 | options: 65 | - Denosing Model Finetuning 66 | - Embeddings Finetuning 67 | - Guidance with Hypernetworks 68 | - Latent Variable Optimization 69 | - Hybrid Finetuning 70 | validations: 71 | required: false 72 | 73 | - type: dropdown 74 | id: training-and-finetuning-free-specifics 75 | attributes: 76 | label: Training and Finetuning Free Specifics (Optional) 77 | description: "If you select 'Training and Finetuning Free', please select a more specific category. If you prefer not to specify, it's okay; we will classify it for you." 78 | options: 79 | - Input Text Refinement 80 | - Inversion/Sampling Modification 81 | - Attention Modification 82 | - Mask Guidance 83 | - Multi-Noise Redirection 84 | validations: 85 | required: false 86 | 87 | - type: textarea 88 | id: additional-information 89 | attributes: 90 | label: Additional Information (Optional) 91 | description: "If there is anything else you'd like to add or specify about your submission, please do so here." 92 | placeholder: "Any other details or context for your paper submission." 93 | validations: 94 | required: false 95 | -------------------------------------------------------------------------------- /.github/workflows/main.yml: -------------------------------------------------------------------------------- 1 | name: Auto Label and Assign Based on Main Method Category 2 | 3 | on: 4 | issues: 5 | types: [opened, edited] 6 | 7 | jobs: 8 | auto-label-and-assign: 9 | runs-on: ubuntu-latest 10 | steps: 11 | - name: Label and Assign Based on Category 12 | uses: actions/github-script@v7 13 | with: 14 | script: | 15 | const issueBody = context.payload.issue.body; 16 | const categories = { 17 | 'Training-Based': { 18 | label: 'training', 19 | assignees: ['Zwette', 'jiaxilv', 'Easquel'] 20 | }, 21 | 'Testing-Time Finetuning': { 22 | label: 'testing-time finetuning', 23 | assignees: ['liuyifan6613', 'Easquel', 'Zwette'] 24 | }, 25 | 'Training and Finetuning Free': { 26 | label: 'training & finetuning free', 27 | assignees: ['huangjch526', 'MingfuYAN', 'Easquel', 'Zwette'] 28 | } 29 | }; 30 | 31 | Object.keys(categories).forEach(category => { 32 | const pattern = new RegExp(`Main Method Category\\s*\\n\\s*${category}`); 33 | if (pattern.test(issueBody)) { 34 | const { label, assignees } = categories[category]; 35 | github.rest.issues.addLabels({ 36 | owner: context.repo.owner, 37 | repo: context.repo.repo, 38 | issue_number: context.issue.number, 39 | labels: [label], 40 | }); 41 | github.rest.issues.addAssignees({ 42 | owner: context.repo.owner, 43 | repo: context.repo.repo, 44 | issue_number: context.issue.number, 45 | assignees: assignees, 46 | }); 47 | } 48 | }); 49 | -------------------------------------------------------------------------------- /EditEval_v1/Dataset/action_change/1.jpeg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/action_change/1.jpeg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/action_change/2.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/action_change/2.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/action_change/3.jpeg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/action_change/3.jpeg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/action_change/4.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/action_change/4.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/background_change/1.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/background_change/1.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/background_change/2.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/background_change/2.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/background_change/3.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/background_change/3.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/background_change/4.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/background_change/4.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/background_change/5.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/background_change/5.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/background_change/6.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/background_change/6.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/background_change/7.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/background_change/7.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/background_change/8.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/background_change/8.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/editing_prompts_collection.xlsx: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/editing_prompts_collection.xlsx -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_addition/1.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_addition/1.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_addition/2.jpeg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_addition/2.jpeg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_addition/3.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_addition/3.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_addition/4.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_addition/4.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_addition/5.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_addition/5.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_addition/6.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_addition/6.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_addition/7.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_addition/7.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_addition/8.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_addition/8.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_removel/1.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_removel/1.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_removel/2.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_removel/2.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_removel/3.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_removel/3.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_removel/4.jpeg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_removel/4.jpeg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_removel/5.jpeg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_removel/5.jpeg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_removel/6.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_removel/6.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_removel/7.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_removel/7.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_removel/8.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_removel/8.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_replacement/1.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_replacement/1.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_replacement/2.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_replacement/2.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_replacement/3.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_replacement/3.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_replacement/4.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_replacement/4.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_replacement/5.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_replacement/5.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_replacement/6.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_replacement/6.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_replacement/7.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_replacement/7.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/object_replacement/8.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/object_replacement/8.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/style_change/1.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/style_change/1.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/style_change/2.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/style_change/2.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/style_change/3.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/style_change/3.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/style_change/4.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/style_change/4.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/style_change/5.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/style_change/5.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/style_change/6.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/style_change/6.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/style_change/7.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/style_change/7.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/style_change/8.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/style_change/8.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/texture_change/1.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/texture_change/1.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/texture_change/2.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/texture_change/2.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/texture_change/3.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/texture_change/3.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/texture_change/4.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/texture_change/4.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/texture_change/5.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/texture_change/5.jpg -------------------------------------------------------------------------------- /EditEval_v1/Dataset/texture_change/6.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/EditEval_v1/Dataset/texture_change/6.jpg -------------------------------------------------------------------------------- /EditEval_v1/Metric/LMM_Score_GPT4V_Prompt_Template.md: -------------------------------------------------------------------------------- 1 | # LMM Score: GPT-4V Prompt Template 2 | 3 | ## Prompt: 4 | I invite you to participate in an experiment on "Evaluation of Image Editing Performance with Different Methods". Be patient and focus. I will pay you tips according to your performance. Are you ready now? 5 | 6 | ## ChatGPT: 7 | 8 | I'm ready and intrigued! Let's dive into the experiment. How can I assist you with the evaluation? 9 | 10 | ## Prompt: 11 | 12 | Step 1: I will provide a source image, a task indicator meaning which specific task is being performed, and an editing instruction that guides the editing process on the source image. Please first view the images and describe the source image. Then describe the editing process. 13 | 14 | ## ChatGPT: 15 | 16 | Please go ahead and provide the source image, task indicator, and editing instruction. I'll start by describing the source image once I see it. 17 | 18 | ## Prompt: 19 | 20 | 21 | This is the source image. 22 | 23 | Task indicator: Background Change. 24 | 25 | Editing Instruction: Change the snowland to a beach. 26 | 27 | ## ChatGPT: 28 | 29 | The source image features an arctic fox sitting on snow-covered ground. The fox has thick, white fur, which provides camouflage against the snowy backdrop. The scene appears to be in a natural, possibly Arctic, environment, given the animal and the snow. There's a soft glow that suggests either early morning or late afternoon light, and the background is predominantly white due to the snow, with some shades of purple, likely due to the lighting or shadows. 30 | 31 | For the editing process as per your instruction, the snowland in the background would be replaced with a beach setting. This would involve removing the snowy ground and any snow-related features from the background and introducing elements typical of a beach scene, such as sand, perhaps some ocean or sea in the distance, and possibly a blue sky. The lighting and shadows on the fox might also need to be adjusted to match the typical lighting one would find in a beach environment, which tends to be brighter and warmer in tone. 32 | 33 | ## Prompt: 34 | Step 2: Now there are 8 edited images which are edited from the source image by 8 different methods with the same one editing instruction mentioned before. You are required to first pay enough careful examination of the 8 edited images. Then you need to score each one based on each of the four following evaluation factors separately, resulting in 4 sub-scores ranging from 1 to 10, termed $S_{acc}, S_{pre}, S_{qua}, S_{real}$, respectively. A higher score means better. The four factors are: 35 | 36 | 1. Editing Accuracy: Evaluate how closely the edited image adheres to the specified editing instruction, measuring the precision of the editing. 37 | 38 | 2. Contextual Preservation: Assess how well the edited image maintains the context of the source image that should not be changed. 39 | 40 | 3. Visual Quality: Assess the overall visual quality of the edited image. 41 | 42 | 4. Logical Realism: Evaluate how logically realistic the edited image is in terms of adherence to natural physical laws. 43 | 44 | Before scoring, carefully specify what to be concerned in each factor in this case. This can help you score more accurately. After that, I will upload the 8 edited images one by one. When I upload one image each time, you first give a detailed analysis based on the factors. After I upload them all, you can score them for comparison. Are you clear? 45 | 46 | ## ChatGPT: 47 | 48 | Absolutely clear. Before scoring, let's define the criteria for each evaluation factor in the context of changing the background from a snowy landscape to a beach scene: 49 | 50 | 1. **Editing Accuracy ($S_{acc}$):** This assesses how well the beach replaces the snowy environment without affecting the fox. Key concerns would include whether the snow is completely replaced by sand, if the boundary between the fox and the new background is seamless, and if elements such as the ocean, sky, or beach accessories (like seashells, starfish, etc.) have been added appropriately. 51 | 52 | 2. **Contextual Preservation ($S_{pre}$):** This evaluates the maintenance of the source image's context that shouldn't be altered, particularly the fox. Points to check are whether the fox remains unaltered in terms of appearance, pose, and shadow, and if the new background is adjusted to maintain the original lighting and perspective context. 53 | 54 | 3. **Visual Quality ($S_{qua}$):** Here, the overall aesthetic appeal of the image is judged, including the resolution, absence of artifacts, color harmony between the fox and the new background, and the natural integration of the beach scene. 55 | 56 | 4. **Logical Realism ($S_{real}$):** This factor gauges the believability of the edited image. It involves assessing whether the fox looks like it could naturally exist in the beach setting, whether the shadows and lighting on the fox match the new environment, and if the scale and perspective of the beach elements align with the fox. 57 | 58 | I'm ready to analyze the edited images based on these factors. Please upload the images when you are ready. 59 | 60 | ## Prompt: 61 | 62 | 63 | The first edited image. 64 | 65 | ## ChatGPT: 66 | 67 | In the provided edited image, the snowy background has been replaced with what seems to be a sandy beach. The following is a detailed analysis based on the four factors: 68 | 69 | 1. **Editing Accuracy:** The initial instruction was to transition the environment from a wintry snowscape to a coastal beach setting. Here, the original snow has been substituted with sand, which indicates a direct approach to the task. However, this change does not extend to a complete representation of a beach; essential elements like the sea, sky, and possibly vegetation or beach artifacts are missing. The alteration is fundamental but not comprehensive. 70 | 71 | 2. **Contextual Preservation:** The subject of the original image, the arctic fox, has been largely conserved in situ with meticulous attention to its defining characteristics. The preservation here is good as the creature remains unaltered and is the clear focal point of the composition, ensuring that the essence of the original image is intact. 72 | 73 | 3. **Visual Quality:** Assessing the aesthetic merits of the edit, there's a satisfactory level of finesse in blending the fox into its new context. Nevertheless, there is room for enhancement in terms of resolution and clarity to elevate the visual narrative. The current state presents an agreeable visual output, yet it doesn’t quite reach the zenith of high-definition quality. 74 | 75 | 4. **Logical Realism:** The logical realism of this image is challenging. While the fox is well integrated into the scene, the color of the sand doesn't align with natural sand colors, and the lighting has a purple hue that is atypical for a beach setting, making it less believable as a real beach environment. 76 | 77 | I'll hold off on scoring until I've seen all eight images to ensure a fair comparison. Please upload the next image when ready. 78 | 79 | 80 | 81 | ## Prompt: 82 | 83 | 84 | 85 | The second one. 86 | 87 | ## ChatGPT: 88 | 89 | In this second edited image, the arctic fox is placed against a background that seems to be a dusk or dawn beach scene. Here’s the analysis based on the reference points provided: 90 | 91 | 1. **Editing Accuracy:** The sandy beach setting is apparent, and there is a hint of the ocean in the background. The edit adheres to the instruction to change the snowland to a beach, and this is achieved with a noticeable level of detail. 92 | 93 | 2. **Contextual Preservation:** The integrity of the fox is maintained with no apparent distortion. However, there is a slight alteration in the shadow under the fox, which seems to be darker and more pronounced than in the original image. The fur texture and the overall posture of the fox remain consistent with the source image. 94 | 95 | 3. **Visual Quality:** This image displays an increase in visual quality. The blending of the fox into the sandy beach is executed with a higher degree of finesse, and the image is free from any noticeable editing artifacts. The colors are vibrant and draw the viewer's eye. 96 | 97 | 4. **Logical Realism:** The setting sun gives the beach a purple hue that complements the fox's fur, and the shadowing beneath the fox is consistent with this lighting. The positioning of the fox and the inclusion of the ocean add to the image's realistic portrayal of a beach scene. 98 | 99 | The overall appearance of this edit is of a higher quality, taking into account the effort to create a believable beach environment around the arctic fox. The elements combined here result in a coherent and aesthetically pleasing image. Please upload the next image for analysis. 100 | 101 | 102 | 103 | ## Prompt: 104 | 105 | 106 | 107 | The third one. 108 | 109 | ## ChatGPT: 110 | 111 | In the third edited image, the fox is seated on a surface that resembles beach sand, with the ocean visible in the background. Here's an analysis based on the evaluation factors: 112 | 113 | 1. **Editing Accuracy ($S\_{acc}$)**: The editing reflects the task instruction, replacing the snow with a sandy beach. The ocean is present, and the sand has a natural color, which aligns with the requested beach setting. 114 | 2. **Contextual Preservation ($S\_{pre}$)**: The fox is well-preserved with no noticeable distortion, maintaining the source context. However, the shadow cast by the fox appears slightly off in direction and intensity when compared to the original image's lighting context. 115 | 3. **Visual Quality ($S\_{qua}$)**: The image maintains a high level of visual quality. The sand's texture is distinct, and the fox is clear and sharply defined. The background and foreground are balanced, contributing to a high-quality visual appearance. 116 | 4. **Logical Realism ($S\_{real}$)**: This image has an improved sense of realism. The beach environment is believable, and the fox's integration into this setting is convincing. The shadow, while slightly off, does not severely detract from the realism. The overall lighting still carries a purple hue, which affects the natural appearance of the scene. 117 | 118 | Awaiting the remaining images for final scoring. Please upload the next image when ready. 119 | 120 | 121 | 122 | 123 | 124 | 125 | 126 | 127 | 128 | 129 | 130 | 131 | 132 | ## Prompt: 133 | 134 | 135 | 136 | The fourth one. 137 | 138 | ## ChatGPT: 139 | 140 | For the fourth edited image, the arctic fox is set against a beach scene that appears to be taken at a similar time of day as the previous images, with a gentle purple cast to the light: 141 | 142 | 1. **Editing Accuracy:** The transformation of the snow to sand is convincingly done, suggesting a careful application of the editing instruction. The ocean in the background, while minimal, does contribute to the beach scene. 143 | 144 | 2. **Contextual Preservation:** The fox's appearance is consistent with the original image. The fur texture and coloration appear unchanged, preserving the context. The shadow under the fox, however, seems a bit mismatched both in terms of direction and intensity, suggesting an alteration that does not completely align with the original lighting. 145 | 146 | 3. **Visual Quality:** The image presents a clear and crisp depiction of the fox. The sand texture is well defined, and the composition is visually appealing. However, the purple cast over the image affects the natural color balance one would expect in a beach setting. 147 | 148 | 4. **Logical Realism:** The realism is somewhat compromised by the purplish hue, which is not typical for a sunny beach scene. Although the fox is well-placed in the sand, the shadow's direction and the overall color cast may reduce the logical realism of the scene. 149 | 150 | The edit represents a competent blend of the source image with the new beach background, achieving a harmonious and believable scene. Please continue with the next image. 151 | 152 | 153 | 154 | 155 | 156 | 157 | 158 | 159 | 160 | 161 | 162 | ## Prompt: 163 | 164 | 165 | The fifth one. 166 | 167 | ## ChatGPT: 168 | 169 | The fifth image displays the arctic fox sitting on a surface that, upon initial glance, seems to be a sandy beach. However, a detailed look reveals some incongruities: 170 | 171 | 1. **Editing Accuracy:** The foreground where the fox sits has the texture of sand, matching the task of changing the snow to a beach. Yet, the background, with its light blue coloration, is reminiscent of snow rather than the typical brown or golden sands of a beach. 172 | 173 | 2. **Contextual Preservation:** The fox itself is untouched and retains its natural state from the original image. The integrity of the animal’s image within this new context is commendable. 174 | 175 | 3. **Visual Quality:** The image quality around the fox is quite good, with the fur texture sharply contrasting against the background. Nonetheless, the background's resemblance to snow rather than sand somewhat diminishes the visual coherence of a beach scene. 176 | 177 | 4. **Logical Realism:** While the foreground does suggest a beach setting, the background's light blue hue disrupts the logical realism. The setting behind the fox lacks the definitive characteristics of a beach, potentially confusing the viewer's sense of place. 178 | 179 | In essence, while the foreground aligns well with the intended beach setting, the background introduces elements that detract from the beach atmosphere, creating a dissonance that impacts the overall realism of the scene. Please present the next image. 180 | 181 | 182 | 183 | 184 | 185 | 186 | 187 | 188 | 189 | 190 | 191 | 192 | 193 | ## Prompt: 194 | 195 | 196 | 197 | The sixth one. 198 | 199 | ## ChatGPT: 200 | 201 | In this sixth edited image, there is a significant departure from the original source in terms of the background as well as the fox: 202 | 203 | 1. **Editing Accuracy:** The edit does not accurately reflect the task of changing the snowy background to a beach. Instead, the background remains snow-like in color and texture, and the fox has been altered dramatically, now resembling more of an abstract or surreal figure with wing-like extensions. 204 | 205 | 2. **Contextual Preservation:** The context of the original image has not been preserved. The fox's natural appearance has been heavily modified, losing the authenticity of the original image and the natural portrayal of the animal. 206 | 207 | 3. **Visual Quality:** The visual quality of the image has been compromised. The editing introduces elements that are quite abstract, and the integration between the fox and these elements does not provide a realistic or high-quality visual representation. 208 | 209 | 4. **Logical Realism:** The image does not convey a sense of logical realism. The alterations to the fox create a scene that defies natural physical laws, and the background, retaining its snow-like appearance, does not support the intended beach setting. 210 | 211 | The image presents as more of an artistic or conceptual creation rather than a realistic edit of the source image. Please continue with the next image. 212 | 213 | 214 | 215 | 216 | 217 | 218 | 219 | 220 | 221 | 222 | 223 | ## Prompt: 224 | 225 | 226 | The seventh one. 227 | 228 | ## ChatGPT: 229 | 230 | In the seventh edited image, we see the arctic fox placed on what seems to be a sandy surface, with the coloration of the scene reflecting a twilight or dawn atmosphere: 231 | 232 | 1. **Editing Accuracy:** The sandy texture suggests a beach environment and follows the instruction to replace the snowy background. However, beyond the sand, there are not many details that would further define the beach setting. 233 | 234 | 2. **Contextual Preservation:** The original image's context has been maintained well with the fox unchanged in its posture and expression. The natural look of the fox within the sandy setting has been respected, though the background lacks other beach elements that might solidify the context. 235 | 236 | 3. **Visual Quality:** The visual quality is fair, with a clean blend between the fox and its sandy surrounding. The texture of the sand is evident, and there is a decent level of detail that makes the image visually appealing. 237 | 238 | 4. **Logical Realism:** The scene achieves a certain level of realism, as the lighting on the fox and the shadows cast seem plausible within a beach-like setting, although the lighting might not be as bright as one would expect on a typical beach day. 239 | 240 | Overall, the image carries a sense of consistency in its editing, presenting the fox in a new environment that somewhat aligns with the intended beach backdrop. Please present the last image for analysis. 241 | 242 | 243 | 244 | 245 | 246 | 247 | 248 | 249 | 250 | 251 | 252 | 253 | 254 | ## Prompt: 255 | 256 | 257 | The last one. 258 | 259 | ## ChatGPT: 260 | 261 | In the eighth edited image, the arctic fox is presented on a sandy beach with visible ocean waves in the background: 262 | 263 | 1. **Editing Accuracy:** The edit successfully replaces the snow with sand and introduces the ocean, which is in line with the editing instructions. The addition of the ocean is a notable improvement in creating a convincing beach setting. 264 | 265 | 2. **Contextual Preservation:** The fox is preserved well, maintaining its original pose and characteristics. The editing has respected the original image, keeping the focus on the fox while changing the background. 266 | 267 | 3. **Visual Quality:** The quality of the editing is good, though the image seems a bit grainy, possibly due to the sand's texture or the lighting. The fox is integrated well into the beach setting, but there's a slight mismatch in clarity between the fox and the background. 268 | 269 | 4. **Logical Realism:** The realism is enhanced by the addition of the ocean and the appropriate scale of the waves in relation to the fox. The lighting on the fox appears mostly consistent with the beach environment, even though the light seems a bit muted for a beach setting. 270 | 271 | This image provides a credible portrayal of the fox in a beach environment, with improvements in the background elements adding to the realism of the setting. 272 | 273 | Having now analyzed all eight images, we can proceed with the next step of your experiment. 274 | 275 | 276 | 277 | 278 | 279 | 280 | 281 | 282 | 283 | 284 | 285 | ## Prompt: 286 | 287 | Score, please. Please be careful about these factors, which require fine-grained observation. Avoid the same scores. Now go ahead! 288 | 289 | ## ChatGPT: 290 | 291 | Understood, here are the scores for all eight images with short explanations: 292 | 293 | **Image 1:** 294 | 295 | * $S_{acc}$: 6 The snow is replaced by sand, but without additional beach elements. 296 | * $S_{pre}$: 7 The fox is well preserved in its natural state. 297 | * $S_{qua}$: 5 The image quality is decent but could be improved for a clearer beach scene. 298 | * $S_{real}$: 6 The scene is somewhat realistic but lacks some beach-specific details. 299 | 300 | **Image 2:** 301 | 302 | * $S_{acc}$: 6 Sand is present, but the beach setting lacks completeness. 303 | * $S_{pre}$: 8 The fox's context and environment are well maintained. 304 | * $S_{qua}$: 7 The image has a good visual appeal, though it lacks some finer details. 305 | * $S_{real}$: 6 The lighting on the fox is slightly off for a beach setting. 306 | 307 | **Image 3:** 308 | 309 | * $S_{acc}$: 8 The beach background is clear and aligns well with the editing instructions. 310 | * $S_{pre}$: 9 Excellent contextual preservation of the fox. 311 | * $S_{qua}$: 9 High visual quality with a clean and natural integration. 312 | * $S_{real}$: 8 Very believable beach scene, though the lighting could be more vibrant. 313 | 314 | **Image 4:** 315 | 316 | * $S_{acc}$: 7 The image adheres to the task with a clear beach setting. 317 | * $S_{pre}$: 7 The fox is unchanged and fits well into the scene. 318 | * $S_{qua}$: 7 Good overall quality with no editing artifacts. 319 | * $S_{real}$: 7 The image is quite believable, although the lighting could be adjusted for consistency. 320 | 321 | **Image 5:** 322 | 323 | * $S_{acc}$: 6 The beach setting is visible, but the back looks more like snow than sand. 324 | * $S_{pre}$: 10 The fox is perfectly preserved in detail. 325 | * $S_{qua}$: 7 Good visual quality, but the background doesn't fully support the beach theme. 326 | * $S_{real}$: 6 The setting is not fully convincing as a beach scene. 327 | 328 | **Image 6:** 329 | 330 | * $S_{acc}$: 1 The edit strays significantly from a beach setting. 331 | * $S_{pre}$: 2 The fox is heavily altered, losing the original context. 332 | * $S_{qua}$: 1 Low visual quality, with the image taking on an abstract quality. 333 | * $S_{real}$: 1 The image lacks realism, with surreal modifications. 334 | 335 | **Image 7:** 336 | 337 | * $S_{acc}$: 7 Beach sand is evident, though additional details are missing. 338 | * $S_{pre}$: 8 The fox remains well-preserved in its original form. 339 | * $S_{qua}$: 7 The image is visually coherent, despite some graininess. 340 | * $S_{real}$: 6 Realism is present but slightly diminished by the muted lighting. 341 | 342 | **Image 8:** 343 | 344 | * $S_{acc}$: 9 The image successfully introduces a clear beach background with sand and sea. 345 | * $S_{pre}$: 9 The fox is excellently maintained in its original state within the new setting. 346 | * $S_{qua}$: 6 The overall quality is good, with minor issues in graininess and clarity. 347 | * $S_{real}$: 7 The setting is realistic, with a few minor discrepancies in lighting. 348 | 349 | These scores are assigned with careful consideration of the nuanced differences between the images based on the provided reference points. 350 | 351 | ## Prompt: 352 | 353 | Step 3: Once you produce the four sub-scores for each edited image, you also need to calculate an overall score $S_{all}$ through a weighted sum of them according to the formula below: 354 | 355 | $S_{all}=0.4S_{acc}+0.3S_{pre}+0.2S_{qua}+0.1S_{real}$ 356 | 357 | ## ChatGPT: 358 | 359 | The overall scores \( $S_{all}$ \) for all eight edited images, calculated using the weighted sum formula, are as follows: 360 | 361 | - Image 1: 6.1 362 | - Image 2: 6.8 363 | - Image 3: 8.5 364 | - Image 4: 7.0 365 | - Image 5: 7.4 366 | - Image 6: 1.3 367 | - Image 7: 7.2 368 | - Image 8: 8.2 369 | 370 | These scores are based on individual assessments of editing accuracy, contextual preservation, visual quality, and logical realism for each image. Based on these scores, Image 3 achieved the highest overall score, indicating it most effectively met the criteria of the editing task. 371 | -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- 1 | MIT License 2 | 3 | Copyright (c) 2024 SiatMMLab 4 | 5 | Permission is hereby granted, free of charge, to any person obtaining a copy 6 | of this software and associated documentation files (the "Software"), to deal 7 | in the Software without restriction, including without limitation the rights 8 | to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 9 | copies of the Software, and to permit persons to whom the Software is 10 | furnished to do so, subject to the following conditions: 11 | 12 | The above copyright notice and this permission notice shall be included in all 13 | copies or substantial portions of the Software. 14 | 15 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 16 | IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 17 | FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE 18 | AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 19 | LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, 20 | OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE 21 | SOFTWARE. 22 | -------------------------------------------------------------------------------- /Leaderboard.md: -------------------------------------------------------------------------------- 1 | # Leaderboard 2 | 3 | ## Table of Contents 4 | - [Object Addition](#object-addition) 5 | - [Object Replacement](#object-replacement) 6 | - [Object Removal](#object-removal) 7 | - [Background Change](#background-change) 8 | - [Style Change](#style-change) 9 | - [Texture Change](#texture-change) 10 | - [Action Change](#action-change) 11 | 12 | 13 | ## Object Addition 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | 32 | 33 | 34 | 35 | 36 | 37 | 38 | 39 | 40 | 41 | 42 | 43 | 44 | 45 | 46 | 47 | 48 | 49 | 50 | 51 | 52 | 53 | 54 | 55 | 56 | 57 | 58 | 59 | 60 | 61 | 62 | 63 | 64 | 65 | 66 | 67 | 68 | 69 | 70 | 71 | 72 | 73 | 74 | 75 | 76 | 77 | 78 | 79 | 80 | 81 | 82 | 83 | 84 | 85 | 86 | 87 | 88 | 89 | 90 |
RankModelPublicationDatasetLMM Score
Mean↑Standard Deviation ↓
1ImagicImagic: Text-based real image editing with diffusion models (CVPR 2023)EditEval_v17.801.27
2InstructDiffusionInstructDiffusion: A generalist modeling interface for vision tasks (CVPR 2024)EditEval_v17.591.89
3ProxEditProxEdit: Improving Tuning-Free Real Image Editing With Proximal Guidance (WACV 2024)EditEval_v17.061.53
4InstructPix2PixInstructPix2Pix: Learning to follow image editing instructions (CVPR 2023)EditEval_v16.882.31
5AIDIEffective real image editing with accelerated iterative diffusion inversion (ICCV 2023)EditEval_v16.811.79
6DDPM InversionAn edit friendly ddpm noise space: Inversion and manipulations (CVPR 2024)EditEval_v16.761.69
7LEDITS++LEDITS++: Limitless Image Editing using Text-to-Image Models (CVPR 2024)EditEval_v16.741.72
8DiffusionDisentanglementUncovering the disentanglement capability in text-to-image diffusion models (CVPR 2023)EditEval_v16.141.74
91 | 92 | ## Object Replacement 93 | 94 | 95 | 96 | 97 | 98 | 99 | 100 | 101 | 102 | 103 | 104 | 105 | 106 | 107 | 108 | 109 | 110 | 111 | 112 | 113 | 114 | 115 | 116 | 117 | 118 | 119 | 120 | 121 | 122 | 123 | 124 | 125 | 126 | 127 | 128 | 129 | 130 | 131 | 132 | 133 | 134 | 135 | 136 | 137 | 138 | 139 | 140 | 141 | 142 | 143 | 144 | 145 | 146 | 147 | 148 | 149 | 150 | 151 | 152 | 153 | 154 | 155 | 156 | 157 | 158 | 159 | 160 | 161 | 162 | 163 | 164 | 165 | 166 | 167 | 168 | 169 | 170 |
RankModelPublicationDatasetLMM Score
Mean↑Standard Deviation ↓
1Null-Text InversionNull-text inversion for editing real images using guided diffusion models (CVPR 2023)EditEval_v18.150.80
2AIDIEffective real image editing with accelerated iterative diffusion inversion (ICCV 2023)EditEval_v17.841.00
3DiffusionDisentanglementUncovering the disentanglement capability in text-to-image diffusion models (CVPR 2023)EditEval_v17.661.41
4ProxEditProxEdit: Improving Tuning-Free Real Image Editing With Proximal Guidance (WACV 2024)EditEval_v17.531.63
5LEDITS++LEDITS++: Limitless Image Editing using Text-to-Image Models (CVPR 2024)EditEval_v17.411.86
6ImagicImagic: Text-based real image editing with diffusion models (CVPR 2023)EditEval_v17.221.65
7InstructDiffusionInstructDiffusion: A generalist modeling interface for vision tasks (CVPR 2024)EditEval_v16.551.46
8InstructPix2PixInstructPix2Pix: Learning to follow image editing instructions (CVPR 2023)EditEval_v15.001.95
171 | 172 | ## Object Removal 173 | 174 | 175 | 176 | 177 | 178 | 179 | 180 | 181 | 182 | 183 | 184 | 185 | 186 | 187 | 188 | 189 | 190 | 191 | 192 | 193 | 194 | 195 | 196 | 197 | 198 | 199 | 200 | 201 | 202 | 203 | 204 | 205 | 206 | 207 | 208 | 209 | 210 | 211 | 212 | 213 | 214 | 215 | 216 | 217 | 218 | 219 | 220 | 221 | 222 | 223 | 224 | 225 |
RankModelPublicationDatasetLMM Score
Mean↑Standard Deviation ↓
1LEDITS++LEDITS++: Limitless Image Editing using Text-to-Image Models (CVPR 2024)EditEval_v18.651.29
2ProxEditProxEdit: Improving Tuning-Free Real Image Editing With Proximal Guidance (WACV 2024)EditEval_v17.751.26
3InstructDiffusionInstructDiffusion: A generalist modeling interface for vision tasks (CVPR 2024)EditEval_v17.481.68
4AIDIEffective real image editing with accelerated iterative diffusion inversion (ICCV 2023)EditEval_v16.721.53
5Inst-InpaintInst-Inpaint: Instructing to Remove Objects with Diffusion Models (arXiv 2023)EditEval_v16.101.54
226 | 227 | ## Background Change 228 | 229 | 230 | 231 | 232 | 233 | 234 | 235 | 236 | 237 | 238 | 239 | 240 | 241 | 242 | 243 | 244 | 245 | 246 | 247 | 248 | 249 | 250 | 251 | 252 | 253 | 254 | 255 | 256 | 257 | 258 | 259 | 260 | 261 | 262 | 263 | 264 | 265 | 266 | 267 | 268 | 269 | 270 | 271 | 272 | 273 | 274 | 275 | 276 | 277 | 278 | 279 | 280 | 281 | 282 | 283 | 284 | 285 | 286 | 287 | 288 | 289 | 290 | 291 | 292 | 293 | 294 | 295 | 296 | 297 | 298 | 299 | 300 | 301 | 302 | 303 | 304 | 305 |
RankModelPublicationDatasetLMM Score
Mean↑Standard Deviation ↓
1Null-Text InversionNull-text inversion for editing real images using guided diffusion models (CVPR 2023)EditEval_v17.281.17
2LEDITS++LEDITS++: Limitless Image Editing using Text-to-Image Models (CVPR 2024)EditEval_v16.910.97
4InstructPix2PixInstructPix2Pix: Learning to follow image editing instructions (CVPR 2023)EditEval_v16.512.49
3ProxEditProxEdit: Improving Tuning-Free Real Image Editing With Proximal Guidance (WACV 2024)EditEval_v16.350.78
6DDSDelta denoising score (ICCV 2023)EditEval_v16.170.81
7AIDIEffective real image editing with accelerated iterative diffusion inversion (ICCV 2023)EditEval_v15.952.34
8EDICTEDICT: Efficient Diffusion for Image Composition and Transformation (CVPR 2023)EditEval_v15.011.94
5SINESine: Single image editing with text-to-image diffusion models (CVPR 2023)EditEval_v13.021.83
306 | 307 | ## Style Change 308 | 309 | 310 | 311 | 312 | 313 | 314 | 315 | 316 | 317 | 318 | 319 | 320 | 321 | 322 | 323 | 324 | 325 | 326 | 327 | 328 | 329 | 330 | 331 | 332 | 333 | 334 | 335 | 336 | 337 | 338 | 339 | 340 | 341 | 342 | 343 | 344 | 345 | 346 | 347 | 348 | 349 | 350 | 351 | 352 | 353 | 354 | 355 | 356 | 357 | 358 | 359 | 360 | 361 | 362 | 363 | 364 | 365 | 366 | 367 | 368 | 369 | 370 | 371 | 372 | 373 | 374 | 375 | 376 | 377 | 378 | 379 | 380 | 381 | 382 | 383 | 384 | 385 |
RankModelPublicationDatasetLMM Score
Mean↑Standard Deviation ↓
1InstructPix2PixInstructPix2Pix: Learning to follow image editing instructions (CVPR 2023)EditEval_v18.210.40
2InstructDiffusionInstructDiffusion: A generalist modeling interface for vision tasks (CVPR 2024)EditEval_v17.410.66
3ImagicImagic: Text-based real image editing with diffusion models (CVPR 2023)EditEval_v17.261.25
4Null-Text InversionNull-text inversion for editing real images using guided diffusion models (CVPR 2023)EditEval_v17.160.90
5DDPM InversionAn edit friendly ddpm noise space: Inversion and manipulations (CVPR 2024)EditEval_v17.080.81
6AIDIEffective real image editing with accelerated iterative diffusion inversion (ICCV 2023)EditEval_v16.991.07
8LEDITS++LEDITS++: Limitless Image Editing using Text-to-Image Models (CVPR 2024)EditEval_v16.861.20
7ProxEditProxEdit: Improving Tuning-Free Real Image Editing With Proximal Guidance (WACV 2024)EditEval_v16.801.07
386 | 387 | ## Texture Change 388 | 389 | 390 | 391 | 392 | 393 | 394 | 395 | 396 | 397 | 398 | 399 | 400 | 401 | 402 | 403 | 404 | 405 | 406 | 407 | 408 | 409 | 410 | 411 | 412 | 413 | 414 | 415 | 416 | 417 | 418 | 419 | 420 | 421 | 422 | 423 | 424 | 425 | 426 | 427 | 428 | 429 | 430 | 431 | 432 | 433 | 434 | 435 | 436 | 437 | 438 | 439 | 440 | 441 | 442 | 443 | 444 | 445 | 446 | 447 | 448 | 449 | 450 | 451 | 452 | 453 | 454 | 455 | 456 | 457 | 458 | 459 | 460 | 461 | 462 | 463 | 464 |
RankModelPublicationDatasetLMM Score
Mean↑Standard Deviation ↓
1InstructDiffusionInstructDiffusion: A generalist modeling interface for vision tasks (CVPR 2024)EditEval_v17.131.83
2Null-Text InversionNull-text inversion for editing real images using guided diffusion models (CVPR 2023)EditEval_v16.821.44
3DiffusionDisentanglementUncovering the Disentanglement Capability in Text-to-Image Diffusion Models (CVPR 2023)EditEval_v16.781.07
4PnPPlug-and-play diffusion features for text-driven image-to-image translation (CVPR 2023)EditEval_v16.631.18
5CycleDiffusionA latent space of stochastic diffusion models for zero-shot image editing and guidance (ICCV 2023)EditEval_v16.551.00
6InstructPix2PixInstructPix2Pix: Learning to follow image editing instructions (CVPR 2023)EditEval_v16.101.41
7Energy-Based Cross AttentionEnergy-Based Cross Attention for Bayesian Context Update in Text-to-Image Diffusion Models (NeurIPS 2023)EditEval_v15.980.75
8ImagicImagic: Text-based real image editing with diffusion models (CVPR 2023)EditEval_v15.571.49
465 | 466 | ## Action Change 467 | 468 | 469 | 470 | 471 | 472 | 473 | 474 | 475 | 476 | 477 | 478 | 479 | 480 | 481 | 482 | 483 | 484 | 485 | 486 | 487 | 488 | 489 | 490 | 491 | 492 | 493 | 494 | 495 | 496 | 497 | 498 | 499 | 500 | 501 | 502 | 503 | 504 | 505 | 506 | 507 | 508 | 509 | 510 | 511 | 512 |
RankModelPublicationDatasetLMM Score
Mean↑Standard Deviation ↓
1ImagicImagic: Text-based real image editing with diffusion models (CVPR 2023)EditEval_v16.970.80
2ForgeditForgedit: Text Guided Image Editing via Learning and Forgetting (arXiv 2023)EditEval_v16.380.87
3MasaCtrlMasaCtrl: Tuning-Free Mutual Self-Attention Control for Consistent Image Synthesis and Editing (ICCV 2023)EditEval_v16.551.16
4TICTuning-Free Inversion-Enhanced Control for Consistent Image Editing (arXiv 2023)EditEval_v16.201.70
513 | -------------------------------------------------------------------------------- /README.MD: -------------------------------------------------------------------------------- 1 | [//]: # (# Diffusion Models for Image Editing) 2 | 3 |

4 | image 5 |

6 | 7 | [![Awesome](media/badge.svg)](https://github.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods) 8 | [![License: MIT](media/License-MIT-green.svg)](https://opensource.org/licenses/MIT) 9 | [![Made With Love](media/Made-With-Love-red.svg)](https://github.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods) 10 | [![arXiv](media/arXiv-Paper-red.svg)](https://arxiv.org/abs/2402.17525) 11 | [![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) 12 | 13 | The repository is based on our survey [Diffusion Model-Based Image Editing: A Survey](https://arxiv.org/pdf/2402.17525.pdf) (TPAMI 2025). 14 | 15 | Yi Huang*, Jiancheng Huang*, Yifan Liu*, Mingfu Yan*, Jiaxi Lv*, Jianzhuang Liu*, Wei Xiong, He Zhang, Liangliang Cao, Shifeng Chen 16 | 17 | Shenzhen Institute of Advanced Technology (SIAT), Chinese Academy of Sciences (CAS), Adobe Inc, Apple Inc, Southern University of Science and Technology (SUSTech) 18 | 19 | ## Abstract 20 | Denoising 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 field. 21 | We 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. 22 | In 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. 23 | To further evaluate the performance of text-guided image editing algorithms, we propose a systematic benchmark, EditEval, featuring an innovative metric, LMM Score. 24 | Finally, we address current limitations and envision some potential directions for future research. 25 | 26 | ## 🔖 News!!! 27 | 28 | 📌 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. 29 | 30 | 📰 2025-02-11: 🥳 Congrats, our paper is accepted by TPAMI 2025!! 31 | 32 | 📰 2024-10-25: Our benchmark [EditEval_v2](#benchmark-editeval_v2) is now released. 33 | 34 | 📰 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. 35 | 36 | 📰 2024-03-14: Our benchmark [EditEval_v1](#benchmark-editeval_v1) is now released. 37 | 38 | 📰 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. 39 | 40 | 📰 2024-02-28: Our comprehensive survey paper, summarizing related methods published before February 1, 2024, is now available. 41 | 42 | 43 | 44 | ## 🔍 BibTeX 45 | If you find this work helpful in your research, welcome to cite the paper and give a ⭐. 46 | 47 | ``` 48 | @article{huang2025diffusion, 49 | title={Diffusion Model-Based Image Editing: A Survey}, 50 | 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}, 51 | journal={IEEE Transactions on Pattern Analysis and Machine Intelligence}, 52 | year={2025}, 53 | publisher={IEEE} 54 | } 55 | ``` 56 | 57 | 58 | 59 | ## Table of contents 60 | - [Papers](#papers) 61 | -
62 | Training-Based 63 | 64 | - [Training-Based: Domain-Specific Editing with Weak Supervision](#training-based-domain-specific-editing-with-weak-supervision) 65 | - [Training-Based: Reference and Attribute Guidance via Self-Supervision](#training-based-reference-and-attribute-guidance-via-self-supervision) 66 | - [Training-Based: Instructional Editing via Full Supervision](#training-based-instructional-editing-via-full-supervision) 67 | - [Training-Based: Pseudo-Target Retrieval with Weak Supervision](#training-based-pseudo-target-retrieval-with-weak-supervision) 68 |
69 | -
70 | Testing-Time Finetuning 71 | 72 | - [Testing-Time Finetuning: Denosing Model Finetuning](#testing-time-finetuning-denosing-model-finetuning) 73 | - [Testing-Time Finetuning: Embeddings Finetuning](#testing-time-finetuning-embeddings-finetuning) 74 | - [Testing-Time Finetuning: Guidance with Hypernetworks](#testing-time-finetuning-guidance-with-hypernetworks) 75 | - [Testing-Time Finetuning: Latent Variable Optimization](#testing-time-finetuning-latent-variable-optimization) 76 | - [Testing-Time Finetuning: Hybrid Finetuning](#testing-time-finetuning-hybrid-finetuning) 77 |
78 | -
79 | Training and Finetuning Free 80 | 81 | - [Training and Finetuning Free: Input Text Refinement](#training-and-finetuning-free-input-text-refinement) 82 | - [Training and Finetuning Free: Inversion/Sampling Modification](#training-and-finetuning-free-inversionsampling-modification) 83 | - [Training and Finetuning Free: Attention Modification](#training-and-finetuning-free-attention-modification) 84 | - [Training and Finetuning Free: Mask Guidance](#training-and-finetuning-free-mask-guidance) 85 | - [Training and Finetuning Free: Multi-Noise Redirection](#training-and-finetuning-free-multi-noise-redirection) 86 |
87 | - [Benchmark EditEval_v1](#benchmark-editeval_v1) 88 | - [Template of Computing LMM Score](EditEval_v1/Metric/LMM_Score_GPT4V_Prompt_Template.md) 89 | - [Leaderboard](#leaderboard) 90 | - [Star History](#star-history) 91 | 92 | 93 | # Papers 94 | ## Training-Based 95 | 96 | ### Training-Based: Domain-Specific Editing 97 | | Title | Publication | Date | 98 | |-------------------------------------------------------------------------------------------|--------------------|--------------| 99 | | [TexFit: Text-Driven Fashion Image Editing with Diffusion Models](https://ojs.aaai.org/index.php/AAAI/issue/view/584) | AAAI 2024 | 2024.03 | 100 | | [CycleNet: Rethinking Cycle Consistency in Text-Guided Diffusion for Image Manipulation](https://arxiv.org/abs/2310.13165) | NeurIPS 2023 | 2023.10 | 101 | | [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 | 102 | | [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 | 103 | | [Towards Real-time Text-driven Image Manipulation with Unconditional Diffusion Models](https://arxiv.org/abs/2304.04344) | arXiv 2023 | 2023.04 | 104 | | [Fine-grained Image Editing by Pixel-wise Guidance Using Diffusion Models](https://arxiv.org/pdf/2212.02024.pdf) | CVPR workshop 2023 | 2022.12 | 105 | | [Diffstyler: Controllable dual diffusion for text-driven image stylization](https://arxiv.org/pdf/2211.10682.pdf) | TNNLS 2024 | 2022.11 | 106 | | [Diffusion Models Already Have A Semantic Latent Space](https://arxiv.org/pdf/2210.10960.pdf) | ICLR 2022 | 2022.10 | 107 | | [Egsde: Unpaired image-to-image translation via energy-guided stochastic differential equations](https://arxiv.org/pdf/2207.06635.pdf) | NeurIPS 2022 | 2022.07 | 108 | | [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 | 109 | | [Unit-ddpm: Unpaired image translation with denoising diffusion probabilistic models](https://arxiv.org/pdf/2104.05358.pdf) | arXiv 2021 | 2021.04 | 110 | | [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 | 111 | 112 | 113 | ### Training-Based: Reference and Attribute Guided Editing 114 | 115 | | Title | Publication | Date | 116 | |-------------------------------------------------------------------------------------------|--------------------|--------------| 117 | | [MagicEraser: Erasing Any Objects via Semantics-Aware Control](https://arxiv.org/abs/2410.10207) | ECCV 2024 | 2024.10 | 118 | | [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 | 119 | | [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 | 120 | | [DreamInpainter: Text-Guided Subject-Driven Image Inpainting with Diffusion Models](http://arxiv.org/abs/2312.03771) | arXiv 2023 | 2023.12 | 121 | | [Uni-paint: A Unified Framework for Multimodal Image Inpainting with Pretrained Diffusion Model](https://arxiv.org/abs/2310.07222) | ACM MM 2023 | 2023.10 | 122 | | [Face Aging via Diffusion-based Editing](https://arxiv.org/abs/2309.11321) | BMVC 2023 | 2023.09 | 123 | | [Anydoor: Zero-shot object-level image customization](https://arxiv.org/abs/2307.09481) | CVPR 2024 | 2023.07 | 124 | | [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 | 125 | | [Text-to-image editing by image information removal](https://arxiv.org/abs/2305.17489) | WACV 2024 | 2023.05 | 126 | | [Reference-based Image Composition with Sketch via Structure-aware Diffusion Model](https://arxiv.org/abs/2304.09748) | CVPR workshop 2023 | 2023.04 | 127 | | [PAIR-Diffusion: A Comprehensive Multimodal Object-Level Image Editor](https://arxiv.org/abs/2303.17546) | CVPR 2024 | 2023.03 | 128 | | [Imagen editor and editbench: Advancing and evaluating text-guided image inpainting](https://arxiv.org/abs/2212.06909) | CVPR 2023 | 2022.12 | 129 | | [Smartbrush: Text and shape guided object inpainting with diffusion model](https://arxiv.org/abs/2212.05034) | CVPR 2023 | 2022.12 | 130 | | [ObjectStitch: Object Compositing With Diffusion Model](http://arxiv.org/abs/2212.00932) | CVPR 2023 | 2022.12 | 131 | | [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 | 132 | 133 | ### Training-Based: Instructional Editing 134 | | Title | Publication | Date | 135 | |-------------------------------------------------------------------------------------------|--------------------|--------------| 136 | | [FreeEdit: Mask-free Reference-based Image Editing with Multi-modal Instruction](https://arxiv.org/abs/2409.18071) | arXiv 2024 | 2024.09 | 137 | | [EditWorld: Simulating World Dynamics for Instruction-Following Image Editing](https://arxiv.org/abs/2405.14785) | arXiv 2024 | 2024.05 | 138 | | [InstructGIE: Towards Generalizable Image Editing](https://arxiv.org/abs/2403.05018) | arXiv 2024 | 2024.03 | 139 | | [SmartEdit: Exploring Complex Instruction-based Image Editing with Multimodal Large Language Models](http://arxiv.org/abs/2312.06739) | CVPR 2024 | 2023.12 | 140 | | [InstructAny2Pix: Flexible Visual Editing via Multimodal Instruction Following](http://arxiv.org/abs/2312.06738) | arXiv 2023 | 2023.12 | 141 | | [Focus on Your Instruction: Fine-grained and Multi-instruction Image Editing by Attention Modulation](http://arxiv.org/abs/2312.10113) | CVPR 2024 | 2023.12 | 142 | | [Emu edit: Precise image editing via recognition and generation tasks](http://arxiv.org/abs/2311.10089) | arXiv 2023 | 2023.11 | 143 | | [Guiding instruction-based image editing via multimodal large language models](http://arxiv.org/abs/2309.17102) | ICLR 2024 | 2023.09 | 144 | | [Instructdiffusion: A generalist modeling interface for vision tasks](https://arxiv.org/abs/2309.03895) | CVPR 2024 | 2023.09 | 145 | | [MoEController: Instruction-based Arbitrary Image Manipulation with Mixture-of-Expert Controllers](https://arxiv.org/abs/2309.04372) | arXiv 2023 | 2023.09 | 146 | | [ImageBrush: Learning Visual In-Context Instructions for Exemplar-Based Image Manipulation](http://arxiv.org/abs/2308.00906) | NeurIPS 2023 | 2023.08 | 147 | | [Inst-Inpaint: Instructing to Remove Objects with Diffusion Models](http://arxiv.org/abs/2304.03246) | arXiv 2023 | 2023.04 | 148 | | [HIVE: Harnessing Human Feedback for Instructional Visual Editing](https://arxiv.org/abs/2303.09618) | CVPR 2024 | 2023.03 | 149 | | [DialogPaint: A Dialog-based Image Editing Model](http://arxiv.org/abs/2303.10073) | arXiv 2023 | 2023.01 | 150 | | [Learning to Follow Object-Centric Image Editing Instructions Faithfully](https://aclanthology.org/2023.findings-emnlp.646/) | EMNLP 2023 | 2023.01 | 151 | | [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 | 152 | 153 | ### Training-Based: Pseudo-Target Retrieval-Based Editing 154 | 155 | | Title | Publication | Date | 156 | |-------------------------------------------------------------------------------------------|--------------------|--------------| 157 | | [Text-Driven Image Editing via Learnable Regions](https://arxiv.org/abs/2311.16432) | CVPR 2024 | 2023.11 | 158 | | [iEdit: Localised Text-guided Image Editing with Weak Supervision](https://arxiv.org/abs/2305.05947) | arXiv 2023 | 2023.05 | 159 | | [ChatFace: Chat-Guided Real Face Editing via Diffusion Latent Space Manipulation](https://arxiv.org/abs/2305.14742) | arXiv 2023 | 2023.05 | 160 | 161 | 162 | ## Testing-Time Finetuning 163 | 164 | ### Testing-Time Finetuning: Denosing Model Finetuning 165 | 166 | | Title | Publication | Date | 167 | |-------------------------------------------------------------------------------------------|--------------------|--------------| 168 | | [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 | 169 | | [Custom-edit: Text-guided image editing with customized diffusion models](https://arxiv.org/abs/2305.15779) | CVPR workshop 2023 | 2023.05 | 170 | | [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 | 171 | 172 | 173 | ### Testing-Time Finetuning: Embeddings Finetuning 174 | 175 | | Title | Publication | Date | 176 | |-------------------------------------------------------------------------------------------|--------------------|--------------| 177 | | [Dynamic Prompt Learning: Addressing Cross-Attention Leakage for Text-Based Image Editing](https://openreview.net/forum?id=5UXXhVI08r) | NeurIPS 2023 | 2023.09 | 178 | | [Prompt Tuning Inversion for Text-Driven Image Editing Using Diffusion Models](http://arxiv.org/abs/2305.04441) | ICCV 2023 | 2023.05 | 179 | | [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 | 180 | | [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 | 181 | 182 | 183 | 184 | ### Testing-Time Finetuning: Guidance with Hypernetworks 185 | 186 | | Title | Publication | Date | 187 | |-------------------------------------------------------------------------------------------|--------------------|--------------| 188 | | [StyleDiffusion: Prompt-Embedding Inversion for Text-Based Editing](https://arxiv.org/abs/2303.15649) | arXiv 2023 | 2023.05 | 189 | | [Inversion-based creativity transfer with diffusion models](https://arxiv.org/abs/2211.13203) | CVPR 2023 | 2022.11 | 190 | 191 | 192 | 193 | ### Testing-Time Finetuning: Latent Variable Optimization 194 | 195 | | Title | Publication | Date | 196 | |-------------------------------------------------------------------------------------------|--------------------|--------------| 197 | | [StableDrag: Stable Dragging for Point-based Image Editing](https://arxiv.org/abs/2403.04437) | arXiv 2024 | 2024.03 | 198 | | [FreeDrag: Feature Dragging for Reliable Point-based Image Editing](https://arxiv.org/abs/2307.04684) | CVPR 2024 | 2023.12 | 199 | | [Contrastive Denoising Score for Text-guided Latent Diffusion Image Editing](https://arxiv.org/abs/2311.18608) | CVPR 2024 | 2023.11 | 200 | | [MagicRemover: Tuning-free Text-guided Image inpainting with Diffusion Models](https://arxiv.org/abs/2310.02848) | arXiv 2023 | 2023.10 | 201 | | [Dragondiffusion: Enabling drag-style manipulation on diffusion models](https://arxiv.org/abs/2307.02421) | ICLR 2024 | 2023.07 | 202 | | [DragDiffusion: Harnessing Diffusion Models for Interactive Point-based Image Editing](https://arxiv.org/abs/2306.14435) | CVPR 2024 | 2023.06 | 203 | | [Delta denoising score](https://openaccess.thecvf.com/content/ICCV2023/html/Hertz_Delta_Denoising_Score_ICCV_2023_paper.html) | ICCV 2023 | 2023.04 | 204 | | [Directed Diffusion: Direct Control of Object Placement through Attention Guidance](https://arxiv.org/pdf/2302.13153.pdf) | AAAI 2024 | 2023.02 | 205 | | [Diffusion-based Image Translation using disentangled style and content representation](https://arxiv.org/abs/2209.15264) | ICLR 2022 | 2022.09 | 206 | 207 | 208 | ### Testing-Time Finetuning: Hybrid Finetuning 209 | 210 | | Title | Publication | Date | 211 | |-------------------------------------------------------------------------------------------|--------------------|--------------| 212 | | [Forgedit: Text Guided Image Editing via Learning and Forgetting](https://arxiv.org/abs/2309.10556) | arXiv 2023 | 2023.09 | 213 | | [LayerDiffusion: Layered Controlled Image Editing with Diffusion Models](https://arxiv.org/abs/2305.18676) | arXiv 2023 | 2023.05 | 214 | | [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 | 215 | | [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 | 216 | 217 | 218 | 219 | ## Training and Finetuning Free 220 | 221 | ### Training and Finetuning Free: Input Text Refinement 222 | 223 | | Title | Publication | Date | 224 | |-------------------------------------------------------------------------------------------|--------------------|--------------| 225 | | [User-friendly Image Editing with Minimal Text Input: Leveraging Captioning and Injection Techniques](https://arxiv.org/abs/2306.02717) | arXiv 2023 | 2023.06 | 226 | | [ReGeneration Learning of Diffusion Models with Rich Prompts for Zero-Shot Image Translation](https://arxiv.org/abs/2305.04651) | arXiv 2023 | 2023.05 | 227 | | [InstructEdit: Improving Automatic Masks for Diffusion-based Image Editing With User Instructions](https://arxiv.org/abs/2305.18047) | arXiv 2023 | 2023.05 | 228 | | [Preditor: Text guided image editing with diffusion prior](https://arxiv.org/abs/2302.07979) | arXiv 2023 | 2023.02 | 229 | 230 | ### Training and Finetuning Free: Inversion/Sampling Modification 231 | 232 | | Title | Publication | Date | 233 | |-------------------------------------------------------------------------------------------|--------------------|--------------| 234 | | [FireFlow: Fast Inversion of Rectified Flow for Image Semantic Editing](https://arxiv.org/abs/2412.07517) | arXiv 2024 | 2024.12 | 235 | | [Inversion-Free Image Editing with Natural Language](https://arxiv.org/abs/2312.04965) | CVPR 2024 | 2023.12 | 236 | | [Fixed-point Inversion for Text-to-image diffusion models](https://arxiv.org/abs/2312.12540) | arXiv 2023 | 2023.12 | 237 | | [Tuning-Free Inversion-Enhanced Control for Consistent Image Editing](https://arxiv.org/abs/2312.14611) | arXiv 2023 | 2023.12 | 238 | | [The Blessing of Randomness: SDE Beats ODE in General Diffusion-based Image Editing](https://arxiv.org/abs/2311.01410) | ICLR 2024 | 2023.11 | 239 | | [LEDITS++: Limitless Image Editing using Text-to-Image Models](https://arxiv.org/abs/2311.16711) | CVPR 2024 | 2023.11 | 240 | | [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 | 241 | | [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 | 242 | | [Fec: Three finetuning-free methods to enhance consistency for real image editing](https://arxiv.org/abs/2309.14934) | arXiv 2023 | 2023.09 | 243 | | [Iterative multi-granular image editing using diffusion models](https://arxiv.org/abs/2309.00613) | WACV 2024 | 2023.09 | 244 | | [ProxEdit: Improving Tuning-Free Real Image Editing With Proximal Guidance](https://arxiv.org/abs/2306.05414) | WACV 2024 | 2023.06 | 245 | | [Diffusion self-guidance for controllable image generation](https://arxiv.org/abs/2306.00986) | NeurIPS 2023 | 2023.06 | 246 | | [Diffusion Brush: A Latent Diffusion Model-based Editing Tool for AI-generated Images](https://arxiv.org/abs/2306.00219) | arXiv 2023 | 2023.06 | 247 | | [Null-text guidance in diffusion models is secretly a cartoon-style creator](https://arxiv.org/abs/2305.06710) | ACM MM 2023 | 2023.05 | 248 | | [Negative-prompt Inversion: Fast Image Inversion for Editing with Text-guided Diffusion Models](https://arxiv.org/abs/2305.16807) | arXiv 2023 | 2023.05 | 249 | | [An Edit Friendly DDPM Noise Space: Inversion and Manipulations](https://github.com/inbarhub/DDPM_inversion) | CVPR 2024 | 2023.04 | 250 | | [Training-Free Content Injection Using H-Space in Diffusion Models](https://arxiv.org/abs/2303.15403) | WACV 2024 | 2023.03 | 251 | | [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 | 252 | | [Direct inversion: Optimization-free text-driven real image editing with diffusion models](https://arxiv.org/abs/2211.07825) | arXiv 2022 | 2022.11 | 253 | 254 | 255 | ### Training and Finetuning Free: Attention Modification 256 | | Title | Publication | Date | 257 | |-------------------------------------------------------------------------------------------|--------------------|--------------| 258 | | [KV-Edit: Training-Free Image Editing for Precise Background Preservation](https://arxiv.org/abs/2502.17363) | arXiv 2025 | 2025.02 | 259 | | [LIME: Localized Image Editing via Attention Regularization in Diffusion Models](https://arxiv.org/abs/2312.09256) | WACV 2025 | 2024.12 | 260 | | [Towards Understanding Cross and Self-Attention in Stable Diffusion for Text-Guided Image Editing](https://arxiv.org/abs/2403.03431) | CVPR 2024 | 2024.03 | 261 | | [HD-Painter: High-Resolution and Prompt-Faithful Text-Guided Image Inpainting with Diffusion Models](https://arxiv.org/abs/2312.14091) | arXiv 2023 | 2023.12 | 262 | | [Tf-icon: Diffusion-based training-free cross-domain image composition](https://arxiv.org/abs/2307.12493) | ICCV 2023 | 2023.07 | 263 | | [Energy-Based Cross Attention for Bayesian Context Update in Text-to-Image Diffusion Models](https://arxiv.org/abs/2306.09869) | NeurIPS 2023 | 2023.06 | 264 | | [Conditional Score Guidance for Text-Driven Image-to-Image Translation](https://arxiv.org/abs/2305.18007) | NeurIPS 2023 | 2023.05 | 265 | | [MasaCtrl: Tuning-Free Mutual Self-Attention Control for Consistent Image Synthesis and Editing](https://arxiv.org/abs/2304.08465) | ICCV 2023 | 2023.04 | 266 | | [Localizing Object-level Shape Variations with Text-to-Image Diffusion Models](https://arxiv.org/abs/2303.11306) | ICCV 2023 | 2023.03 | 267 | | [Zero-shot image-to-image translation](https://dl.acm.org/doi/abs/10.1145/3588432.3591513) | ACM SIGGRAPH 2023 | 2023.02 | 268 | | [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 | 269 | | [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 | 270 | | [Prompt-to-prompt image editing with cross attention control](https://openreview.net/forum?id=_CDixzkzeyb) | ICLR 2023 | 2022.08 | 271 | 272 | 273 | ### Training and Finetuning Free: Mask Guidance 274 | 275 | | Title | Publication | Date | 276 | |-------------------------------------------------------------------------------------------|---------------|---------| 277 | | [Grounded-Instruct-Pix2Pix: Improving Instruction Based Image Editing with Automatic Target Grounding](https://ieeexplore.ieee.org/document/10446377) | ICASSP 2024 | 2024.03 | 278 | | [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 | 279 | | [ZONE: Zero-Shot Instruction-Guided Local Editing](https://arxiv.org/abs/2312.16794) | CVPR 2024 | 2023.12 | 280 | | [Watch your steps: Local image and scene editing by text instructions](https://arxiv.org/abs/2308.08947) | arXiv 2023 | 2023.08 | 281 | | [Energy-Based Cross Attention for Bayesian Context Update in Text-to-Image Diffusion Models](https://arxiv.org/abs/2306.09869) | NeurIPS 2023 | 2023.06 | 282 | | [Differential Diffusion: Giving Each Pixel Its Strength](https://arxiv.org/abs/2306.00950) | arXiv 2023 | 2023.06 | 283 | | [PFB-Diff: Progressive Feature Blending Diffusion for Text-driven Image Editing](https://arxiv.org/abs/2306.16894) | arXiv 2023 | 2023.06 | 284 | | [FISEdit: Accelerating Text-to-image Editing via Cache-enabled Sparse Diffusion Inference](https://arxiv.org/abs/2305.17423) | AAAI 2024 | 2023.05 | 285 | | [Inpaint anything: Segment anything meets image inpainting](https://arxiv.org/abs/2304.06790) | arXiv 2023 | 2023.04 | 286 | | [Region-aware diffusion for zero-shot text-driven image editing](https://arxiv.org/abs/2302.11797) | CVM 2023 | 2023.02 | 287 | | [Text-guided mask-free local image retouching](https://ieeexplore.ieee.org/abstract/document/10219704) | ICME 2023 | 2022.12 | 288 | | [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 | 289 | | [DiffEdit: Diffusion-based semantic image editing with mask guidance](https://openreview.net/forum?id=3lge0p5o-M-) | ICLR 2023 | 2022.10 | 290 | | [Blended latent diffusion](https://arxiv.org/abs/2206.02779) | SIGGRAPH 2023 | 2022.06 | 291 | 292 | ### Training and Finetuning Free: Multi-Noise Redirection 293 | 294 | 295 | | Title | Publication | Date | 296 | |-------------------------------------------------------------------------------------------|--------------------|--------------| 297 | | [Object-aware Inversion and Reassembly for Image Editing](https://arxiv.org/abs/2310.12149) | ICLR 2024 | 2023.10 | 298 | | [Ledits: Real image editing with ddpm inversion and semantic guidance](https://arxiv.org/abs/2307.00522) | arXiv 2023 | 2023.07 | 299 | | [Sega: Instructing diffusion using semantic dimensions](https://arxiv.org/abs/2301.12247) | NeurIPS 2023 | 2023.01 | 300 | | [The stable artist: Steering semantics in diffusion latent space](https://arxiv.org/abs/2212.06013) | arXiv 2022 | 2022.12 | 301 | 302 | # Benchmark EditEval_v1 303 | 304 | 305 | **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! 306 | 307 | # Benchmark EditEval_v2 308 | 309 | **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! 310 | 311 | # Leaderboard 312 | 313 | To 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). 314 | 315 | # Star History 316 | 317 | ![Star History Chart](https://api.star-history.com/svg?repos=SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods&type=Date) 318 | -------------------------------------------------------------------------------- /media/License-MIT-green.svg: -------------------------------------------------------------------------------- 1 | License: MITLicenseMIT -------------------------------------------------------------------------------- /media/Made-With-Love-red.svg: -------------------------------------------------------------------------------- 1 | Made With: LoveMade WithLove -------------------------------------------------------------------------------- /media/arXiv-Paper-red.svg: -------------------------------------------------------------------------------- 1 | arXiv: PaperarXivPaper -------------------------------------------------------------------------------- /media/arXiv.svg: -------------------------------------------------------------------------------- 1 | arXivarXiv -------------------------------------------------------------------------------- /media/badge.svg: -------------------------------------------------------------------------------- 1 | 2 | -------------------------------------------------------------------------------- /media/bg_change/6_AIDI.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/media/bg_change/6_AIDI.png -------------------------------------------------------------------------------- /media/bg_change/6_DDS.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/media/bg_change/6_DDS.png -------------------------------------------------------------------------------- /media/bg_change/6_InstructionPix2Pix.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/media/bg_change/6_InstructionPix2Pix.png -------------------------------------------------------------------------------- /media/bg_change/6_LEDITS_PLUSPLUS.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/media/bg_change/6_LEDITS_PLUSPLUS.png -------------------------------------------------------------------------------- /media/bg_change/6_SINE.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/media/bg_change/6_SINE.png -------------------------------------------------------------------------------- /media/bg_change/6_edict.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/media/bg_change/6_edict.png -------------------------------------------------------------------------------- /media/bg_change/6_null_text.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/media/bg_change/6_null_text.png -------------------------------------------------------------------------------- /media/bg_change/6_proxedit.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/media/bg_change/6_proxedit.png -------------------------------------------------------------------------------- /media/bg_change/6_src.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/media/bg_change/6_src.jpg -------------------------------------------------------------------------------- /media/title.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SiatMMLab/Awesome-Diffusion-Model-Based-Image-Editing-Methods/0581cec1a20b89ca0f60e70cd6545704d5600368/media/title.png --------------------------------------------------------------------------------