Ai Image Editing Tools | AI Final Year Project Report

Complete Technical Paper Content — Ai Image Editing Tools

1. Executive Abstract, Problem Formulation & Background

ImgEdify / Awesome-GPT4o-Image-Prompts Star 588 📚 GPT4o Prompts Dictionary | Curated Collection of AI Image Generation Prompts ai image-generation prompts prompt-engineering chatgpt ai-image-generator ai-image-creator aiimagegenerator ai-image-editing ai-logo-maker Updated May 15, 2025 HTML

mf-yang / ecommerce-image-ai-processor Star 167 一款面向电商运营的桌面软件，通过 qwen-image-edit-plus AI 模型批量处理商品图片，实现背景去除、商品合成，并添加自定义背景、边框和文字。 desktop-app python ecommerce template-editor image-composition github-actions background-removal pyqt6 ai-image-editing qwen-image-edit-plus Updated Apr 29, 2026 Python

DenRakEiw / scumble Sponsor Star 156 Free, open-source desktop editor for AI inpainting: select, prompt, get a colour-matched layer. FLUX 3 Image, FLUX.2, GPT Image, Nano Banana, Seedream and Qwen through your API key or your own ComfyUI. Layers, PSD export, MCP. electron desktop-app flux mcp psd image-editing image-editor image-generation inpainting ai-art outpainting stable-diffusion generative-ai comfyui ai-image-editing model-context-protocol mcp-server ai-inpainting flux3 Updated Oct 24, 2026 JavaScript

Free, open-source desktop editor for AI inpainting: select, prompt, get a colour-matched layer. FLUX 3 Image, FLUX.2, GPT Image, Nano Banana, Seedream and Qwen through your API key or your own ComfyUI. Layers, PSD export, MCP.

SnapGenAI / Free-Nano-Banana-Pro-API-Ultimate-AI-Image-Generator Star 83 Free AI Image Generator API — Nano Banana Pro (Gemini 3.0 Pro). Create ultra-realistic images, edit photos, and generate stunning AI visuals in seconds — 100% free, no watermark. The ultimate AI Image solution for creators and developers. python machine-learning computer-vision deep-learning python3 pytorch image-editing generative-model image-generation text-to-image ai-api gemini-api diffusion-models stable-diffusion generative-ai ai-image-generator google-gemini ai-image-editing gemini-3-pro nano-banana-pro Updated Dec 1, 2025

Free AI Image Generator API — Nano Banana Pro (Gemini 3.0 Pro). Create ultra-realistic images, edit photos, and generate stunning AI visuals in seconds — 100% free, no watermark. The ultimate AI Image solution for creators and developers.

SamurAIGPT / muapi-comfyui Star 80 ComfyUI custom nodes for 600+ AI models — Seedance, Kling, Veo3, Flux, HiDream, GPT-image, Imagen4 and more via muapi.ai flux image-generation text-to-image lipsync video-generation muapi image-to-video text-to-video image-upscale stable-diffusion generative-ai comfyui ai-video-generation comfyui-nodes ai-image-editing kling veo3 seedance comfyui-custom-nodes seedance2 Updated Oct 15, 2026 Python

ComfyUI custom nodes for 600+ AI models — Seedance, Kling, Veo3, Flux, HiDream, GPT-image, Imagen4 and more via muapi.ai

2. Related Work, Literature Survey & Algorithmic Foundations

XBastille / DeepFX-Studio Sponsor Star 20 DeepFX Studio represents a comprehensive platform that bridges cutting-edge computer vision research with practical deployment. computer-vision deep-learning image-processing gans inpainting diffusion-models ai-image-editing Updated Jan 4, 2026 Python

DeepFX Studio represents a comprehensive platform that bridges cutting-edge computer vision research with practical deployment.

CoderSwarup / media-flow Star 18 MediaFlow is an all-in-one video transcoder and AI-powered image/video editor. It supports seamless transcoding, background removal, caption generation, smart video cropping, and more. Built with Next.js, AWS, Cloudinary, and Prisma, it simplifies media workflows with scalable cloud storage. nodejs redis aws web-development stripe nextjs postgresql mern cloudinary video-transcoding tailwindcss shadcn-ui ai-image-editing accertinity-ui Updated Dec 21, 2024 TypeScript

MediaFlow is an all-in-one video transcoder and AI-powered image/video editor. It supports seamless transcoding, background removal, caption generation, smart video cropping, and more. Built with Next.js, AWS, Cloudinary, and Prisma, it simplifies media workflows with scalable cloud storage.

kevinshowkat / brood Star 18 Reference-first AI image editing desktop for developers (macOS, Tauri, Rust). desktop-app macos rust image-generation tauri openrouter ai-image-editing multimodal-ai promptless reference-first Updated Aug 21, 2026 Rust

BovineOverlord / Loyal-Bear---The-SynthID-Scrambler Star 17 Automatically remove SynthID from any image ai ai-image ai-image-processing ai-image-editing synthid Updated Aug 22, 2026 Python

AI-Tool-List / AI-Photo-Editor Star 14 AI Photo Editor Navigation - Top 50 photos ai photo-editor image-editing-app ai-top-tools bestofai theresanaiforthat awesome-ai-tools ai-tools-directory ai-tools-list top-ai-tools ai-image-editing vibe-coding nanobanana Updated Sep 25, 2025

FernandoS27 / WhiteoutTex Star 13 A simple tool for converting Blizzard Entertainment texture formats. Generates mipmaps and manages proper display and conversion. warcraft tex converter image-processing diablo dds blp starcraft2 mipmaps-generation ai-image-editing diablo4tool Updated Aug 30, 2026 C++

3. Dataset Ingestion, Preprocessing & Stochastic Augmentations

A simple tool for converting Blizzard Entertainment texture formats. Generates mipmaps and manages proper display and conversion.

coreyrab / statickit Star 11 An open-source AI photo editor with presets, subject swapping, version control, and more. Free to use, just bring your own API keys. image-editing image-editor gemini image-editing-tool ai-image-editing nano-banana-pro Updated Apr 21, 2026 TypeScript

An open-source AI photo editor with presets, subject swapping, version control, and more. Free to use, just bring your own API keys.

cronux-ind / ai-video-generation-workflow Star 11 Generate reliable short finance explainer videos with script, slides, voice, subtitles, and batch-ready rendering in a stable, modular workflow. aws awesome typescript web-development video ai ffmpeg postgresql mern cloudinary image-generation video-editor video-generation prompt-engineering ai-video-generator ai-video prompts-template ai-video-generation ai-image-editing seedance2 Updated Oct 12, 2026 TypeScript

Generate reliable short finance explainer videos with script, slides, voice, subtitles, and batch-ready rendering in a stable, modular workflow.

SamurAIGPT / clearmark-ai Star 11 Open-source AI watermark remover — production-ready Next.js SaaS that removes watermarks, logos, stamps, and text overlays from images using GPT Image 2. Free Watermarkremover.io & HitPaw alternative with Stripe billing, credits, NextAuth, and Prisma. stripe nextjs image-inpainting prisma object-removal image-cleanup tailwindcss saas-boilerplate remove-watermark nextjs-boilerplate nextauth watermark-removal generative-ai ai-watermark-remover ai-image-generator ai-saas ai-image-editing ai-photo-editor photo-cleanup hitpaw-alternative Updated Aug 2, 2026 JavaScript

Open-source AI watermark remover — production-ready Next.js SaaS that removes watermarks, logos, stamps, and text overlays from images using GPT Image 2. Free Watermarkremover.io & HitPaw alternative with Stripe billing, credits, NextAuth, and Prisma.

danielrosehill / Nano-Banana-Desktop Star 8 Vibe coded starter for a Nano Banana inpainting utility / image editor (Gemini API) linux-desktop inpainting gemini-api ai-image-editing nano-banana Updated Nov 1, 2025 Python

4. Neural Network Backbone, Layer Specifications & Loss Formulations

ImMappyJ / ps-ai Star 8 PS-AI — A multi-model AI creative toolkit for Photoshop. Edit, composite, relight, and generate images with Nano Banana, GPT Image, Bailian, and ComfyUI. photoshop image-generation photo-editing image-to-image photoshop-plugin image-blending image-compositing adobe-uxp generative-ai comfyui ai-image-editing gpt-image qwen-image nano-banana ai-relighting Updated Oct 30, 2026 TypeScript

PS-AI — A multi-model AI creative toolkit for Photoshop. Edit, composite, relight, and generate images with Nano Banana, GPT Image, Bailian, and ComfyUI.

manthanank / ai-image-studio-app Sponsor Star 7 AI Image Studio nodejs angular express mongodb expressjs gemini express-js gemini-api ai-image-generation ai-image-generator ai-image-editing Updated Apr 1, 2026 HTML

light-and-ray / image-edit-bot-open-webui Star 7 An closedwebui bot-wrapper around comfyui api, for simple image + prompt workflows bot image-editing comfyui openwebui ai-image-editing qwen-image-edit flux-kontext-dev Updated May 24, 2026 Python

nexscope-ai / nexscope-creative-api Star 5 Official Nexscope Creative API documentation and examples for ecommerce image generation, video generation, editing, virtual try-on, and product marketing. javascript python ecommerce curl rest-api ecommerce-api virtual-try-on background-removal product-photography generative-ai ai-video-generation ai-image-editing image-generation-api video-generation-api creative-api nexscope Updated Aug 24, 2026 Python

Official Nexscope Creative API documentation and examples for ecommerce image generation, video generation, editing, virtual try-on, and product marketing.

howardrock88 / semcanvas-ai Star 5 Semantic AI image editing canvas with local Codex CLI, SAM/FastSAM segmentation, and pluggable image APIs. nodejs javascript macos sam image-editing openai image-generation semantic-segmentation inpainting local-first comfyui fastsam ai-image-editing codex-cli Updated May 13, 2026 JavaScript

Semantic AI image editing canvas with local Codex CLI, SAM/FastSAM segmentation, and pluggable image APIs.

5. Training Protocol, Convergence Behavior & Experimental Evaluation

To associate your repository with the ai-image-editing topic, visit your repo's landing page and select "manage topics."

DeepGen 1.0: A Lightweight Unified Multimodal Model for Advancing Image Generation and Editing Shanghai Innovation Institute, DeepGen Team 🔥 News Feb 13, 2026: We released DeepGen 1.0 , Pre-traning, Supervised Fine-Tuning and Reinforcement Learning checkpoints can be found in Huggingface , support both T2I generation and image editing. Feb 13, 2026: We released the training code support Pre-training, Supervised Fine-Tuning, Reinforcement Learning and evaluation code support wide range of benchmarks. Feb 13, 2026: We released the DeepGen 1.0 technical report on Arxiv ✨ Introduction Broader Scenario and Dimension Coverage We propose DeepGen 1.0, a lightweight unified multimodal model with only 5B parameters (3B VLM + 2B DiT). It integrates five core capabilities: general image generation, general image editing, reasoning image generation, reasoning image editing, and text rendering—within a single model. Across multiple authoritative benchmarks, DeepGen 1.0 is competitive with or surpassing the state-of-the-art unified multimodal models that are 3× to 16× larger, achieving comprehensive performance, demonstrating that massive scaling is not the sole path to high-performance multimodal generation. 🧠 Method Our core observation is that a lightweight model, when empowered by synergistic architecture design and data-centric training strategies, can achieve comprehensive capabilities competitive with or even surpassing much larger counterparts. To overcome the limitations of lightweight models in semantic understanding and fine-grained control, we introduce Stacked Channel Bridging (SCB) , a deep alignment framework that extracts hierarchical features from multiple VLM layers and fuses them with learnable "think tokens" to provide the generative backbone with structured, reasoning-rich guidance. We further design a data-centric training strategy spanning three progressive stages: (1) Alignment Pre-training on large-scale image-text pairs and editing triplets to synchronize VLM and DiT representations, (2) Joint Supervised Fine-tuning on a high-quality mixture of generation, editing, and reasoning tasks to foster omni-capabilities, and (3) Reinforcement Learning with MR-GRPO , which leverages a mixture of reward functions and supervision signals, resulting in substantial gains in generation quality and alignment with human preferences, while maintaining stable training progress and avoiding visual artifacts. 💻 Train & Eval Set up environment git clone https://github.com/deepgenteam/deepgen.git cd deepgen conda create -n deepgen python=3.12 -y conda activate deepgen pip install -r requirements.txt pip install flash_attn==2.8.3 --no-build-isolation pip install xtuner==0.2.0 pip install transformers==4.56.1 pip install triton==2.3.0 pip install -U opencv-python-headless Data Prepare Please See DATA for more details. We provide a detailed description of the data download and usage procedures for both the Pre-traning stage and the Supervised Fine-Tuning stage. Inference Diffusers (click to expand) We provide a diffusers-compatible format at 🤗 deepgenteam/DeepGen-1.0-diffusers . Text-to-Image: import torch from diffusers import DiffusionPipeline pipe = DiffusionPipeline . from_pretrained ( "deepgenteam/DeepGen-1.0-diffusers" , torch_dtype = torch . bfloat16 , trust_remote_code = True , ) pipe . to ( "cuda" ) result = pipe ( prompt = "a photo of a blue pizza and a yellow baseball glove" , height = 512 , width = 512 , num_inference_steps = 50 , guidance_scale = 4.0 , seed = 42 , ) result . images [ 0 ]. save ( "output.png" ) Image Editing: from PIL import Image result = pipe ( prompt = "Place this guitar on a sandy beach with the sunset in the background." , image = Image . open ( "guitar.png" ), negative_prompt = "blurry, low quality, low resolution, distorted, deformed, broken content, missing parts, damaged details, artifacts, glitch, noise, pixelated, grainy, compression artifacts, bad composition, wrong proportion, incomplete editing, unfinished, unedited areas." , height = 512 , width = 512 , num_inference_steps = 50 , guidance_scale = 4.0 , seed = 42 , ) result . images [ 0 ]. save ( "edited.png" ) Please refer to INFERENCE for more details, including the native pipeline usage. Train See TRAIN for more details. We provide a detailed description of the model and training configs for both the Pre-traning stage and the Supervised Fine-Tuning stage. Eval We provide the scripts for evaluating wide range of T2I and Editing benchmarks. Please See EVAL for more details. 📊 Benchmarks 1. General Image Generation Model Params Geneval ↑ DPGBench ↑ UniGenBench ↑ OmniGen2 3B + 4B 0.80 83.57 63.09 BAGEL 14B 0.82 85.10 61.53 X-Omni 7B + 12B 0.83 87.65🥉 53.77 Lumina-DiMOO 8B 0.88🥇 86.04 71.12 Hunyuan-Image-3.0 80B 0.72 86.10 — Qwen-Image 7B + 20B 0.87 🥈 88.32 🥇 78.81 🥇 LongCat-Image 7B + 6B 0.87 🥈 86.80 — Z-Image-Turbo 4B + 6B 0.84 85.15 71.40 GLM-Image 9B + 7B — 84.78 — DeepGen 1.0 (SFT) 3B + 2B 0.86 🥉 87.05 74.18 🥉 DeepGen 1.0 (RL) 3B + 2B 0.87 🥈 87.90 🥈 75.74 🥈 2. General Image Editing Model Params GEdit-EN ↑ ImgEdit ↑ BAGEL 14B 6.52 3.20 Qwen-Image-Edit [2509] 7B + 20B 7.54 🥈 4.35 🥈 LongCat-Image-Edit 7B + 6B 7.60 🥇 4.50 🥇 Mammoth2 8B + 3B + 2B 6.60 4.06 DeepGen 1.0 (SFT) 3B + 2B 7.12 4.09 DeepGen 1.0 (RL) 3B + 2B 7.17 🥉 4.14 🥉 3. Reasoning Image Generation Model Params WISE ↑ T2I-CoREBench ↑ OmniGen2 3B + 4B 0.47 36.1 BAGEL 14B 0.70 🥉 41.1 Hunyuan-Image-3.0 80B 0.57 46.0 Qwen-Image 7B + 20B 0.62 46.3 🥉 LongCat-Image 7B + 6B 0.65 52.2 🥇 Z-Image-Turbo 4B + 6B - 43.7 DeepGen 1.0 (SFT) 3B + 2B 0.72 🥈 45.7 DeepGen 1.0 (RL) 3B + 2B 0.73 🥇 46.5 🥈 4. Reasoning Image Editing Model Params RISE ↑ UniREditBench ↑ OmniGen2 3B + 4B - 43.4 BAGEL 14B 11.9 🥈 51.0 Qwen-Image-Edit [2509] 7B + 20B 8.9 56.5 🥉 DeepGen 1.0 (SFT) 3B + 2B 13.3 🥇 77.5 🥇 DeepGen 1.0 (RL) 3B + 2B 10.8 🥉 75.7 🥈 📧 Contact dywang24@m.fudan.edu.cn , wjqdev@gmail.com 🎨 Quantitative results ⭐ Citation @article { wang2026deepgen , title = { DeepGen 1.0: A Lightweight Unified Multimodal Model for Advancing Image Generation and Editing } , author = { Wang, Dianyi and Li, Ruihang and Han, Feng and Ma, Chaofan and Song, Wei and Wang, Siyuan and Wang, Yibin and Xin, Yi and Liu, Hongjian and Zhang, Zhixiong and others } , journal = { arXiv preprint arXiv:2602.12205 } , year = { 2026 } } 🙏 Acknowledgement The project builds upon the following pioneering works: OpenUni : We thank the OpenUni releasing the elegant and concise code and pretrain dataset. UniPic2-SD3.5M-Kontext-2B : We use UniPic2-SD3.5M-Kontext-2B as our diffusion module, considering its efficiency and strong performance on both t2i and editing. UnifiedReward-Think : We use UnifiedReward-Think as our reward model for RL, considering its strong performance. Qwen2.5 VL : We useQwen2.5 VL-3B as our VLM module, considering its efficiency and strong performance on multimodal understanding abilities. BLIP3-o : We thank the BLIP3-o team for releasing the precious high-quality tuning dataset. OpenGPT-4o-Image : We thank the OpenGPT-4o-Image team for releasing the precious high-quality tuning dataset. ShareGPT-4o-Image : We thank the ShareGPT-4o-Image team for releasing the precious high-quality tuning dataset. Echo-4o : We thank the Echo-4o team for releasing the precious high-quality tuning dataset. OmniGen2 : We thank the OmniGen2 team for releasing the precious high-quality editing tuning dataset and code. Uniworld-V1 : We thank the Uniworld team for releasing the precious high-quality tuning dataset and code. Picobanana : We thank the Picobanana team for releasing the precious high-quality editing tuning dataset. Nano-consist : We thank the Nano-consist team for releasing the precious high-quality editing tuning dataset. NHR-edit : We thank the NHR-edit team for releasing the precious high-quality editing tuning dataset. UniREditBench : We thank the UniREditBench team for releasing the precious high-quality reason-based editing tuning dataset.

Broader Scenario and Dimension Coverage We propose DeepGen 1.0, a lightweight unified multimodal model with only 5B parameters (3B VLM + 2B DiT). It integrates five core capabilities: general image generation, general image editing, reasoning image generation, reasoning image editing, and text rendering—within a single model. Across multiple authoritative benchmarks, DeepGen 1.0 is competitive with or surpassing the state-of-the-art unified multimodal models that are 3× to 16× larger, achieving comprehensive performance, demonstrating that massive scaling is not the sole path to high-performance multimodal generation.

Our core observation is that a lightweight model, when empowered by synergistic architecture design and data-centric training strategies, can achieve comprehensive capabilities competitive with or even surpassing much larger counterparts. To overcome the limitations of lightweight models in semantic understanding and fine-grained control, we introduce Stacked Channel Bridging (SCB) , a deep alignment framework that extracts hierarchical features from multiple VLM layers and fuses them with learnable "think tokens" to provide the generative backbone with structured, reasoning-rich guidance. We further design a data-centric training strategy spanning three progressive stages: (1) Alignment Pre-training on large-scale image-text pairs and editing triplets to synchronize VLM and DiT representations, (2) Joint Supervised Fine-tuning on a high-quality mixture of generation, editing, and reasoning tasks to foster omni-capabilities, and (3) Reinforcement Learning with MR-GRPO , which leverages a mixture of reward functions and supervision signals, resulting in substantial gains in generation quality and alignment with human preferences, while maintaining stable training progress and avoiding visual artifacts.

git clone https://github.com/deepgenteam/deepgen.git cd deepgen conda create -n deepgen python=3.12 -y conda activate deepgen pip install -r requirements.txt pip install flash_attn==2.8.3 --no-build-isolation pip install xtuner==0.2.0 pip install transformers==4.56.1 pip install triton==2.3.0 pip install -U opencv-python-headless

Please See DATA for more details. We provide a detailed description of the data download and usage procedures for both the Pre-traning stage and the Supervised Fine-Tuning stage.

import torch from diffusers import DiffusionPipeline pipe = DiffusionPipeline . from_pretrained ( "deepgenteam/DeepGen-1.0-diffusers" , torch_dtype = torch . bfloat16 , trust_remote_code = True , ) pipe . to ( "cuda" ) result = pipe ( prompt = "a photo of a blue pizza and a yellow baseball glove" , height = 512 , width = 512 , num_inference_steps = 50 , guidance_scale = 4.0 , seed = 42 , ) result . images [ 0 ]. save ( "output.png" )

from PIL import Image result = pipe ( prompt = "Place this guitar on a sandy beach with the sunset in the background." , image = Image . open ( "guitar.png" ), negative_prompt = "blurry, low quality, low resolution, distorted, deformed, broken content, missing parts, damaged details, artifacts, glitch, noise, pixelated, grainy, compression artifacts, bad composition, wrong proportion, incomplete editing, unfinished, unedited areas." , height = 512 , width = 512 , num_inference_steps = 50 , guidance_scale = 4.0 , seed = 42 , ) result . images [ 0 ]. save ( "edited.png" )