generative AI / AIGC computer vision
OpenMMLab MMagic is a PyTorch toolbox for multimodal generative AI (AIGC), offering unified APIs and a model zoo for text-to-image, image/video editing, restoration, inpainting, super-resolution, and frame interpolation.
An earlier or more limited way to do the job. Still the right call when you need something small, proven, or CPU-only.
Choose when you need a standalone, high-quality video frame interpolation model without the full MMagic toolbox.
ECCV2022 - Real-Time Intermediate Flow Estimation for Video Frame Interpolation
Choose for training-free inpainting on stable diffusion models (optionally via ComfyUI) rather than MMagic's broader inpainting zoo.
High quality training free inpaint for every stable diffusion model. Supports ComfyUI
Choose for a robust, large-mask inpainting method as a compact standalone model.
๐ฆ LaMa Image Inpainting, Resolution-robust Large Mask Inpainting with Fourier Convolutions, WACV 2022
Choose for practical face restoration only, when you want a specialized model instead of a general-purpose toolbox.
GFPGAN aims at developing Practical Algorithms for Real-world Face Restoration.
Choose for an instruction-following image editing foundation model rather than a multi-task toolbox.
FireRed-Image-Edit is a powerful image editing foundation model achieving open-source state-of-the-art performance with precise instruction following, high-fidelity generation, superior identity consistency, and seamless multi-element fusion.
Choose for a classic GAN-based deblurring method when you need a single-purpose model.
Image Deblurring using Generative Adversarial Networks
Choose for multilingual text-to-image diffusion generation as a self-contained model.
Kandinsky 2 โ multilingual text2image latent diffusion model
Choose to study or extend the original latent diffusion architecture, not for a production toolbox.
High-Resolution Image Synthesis with Latent Diffusion Models
Choose when you need the specific LTX-Video architecture for text-to-video generation.
Official repository for LTX-Video
Choose for practical general-purpose image/video restoration with a focused, widely used implementation.
Real-ESRGAN aims at developing Practical Algorithms for General Image/Video Restoration.
Choose for text-driven consistency-aware video editing as a standalone diffusion pipeline.
[ICCV 2023] StableVideo: Text-driven Consistency-aware Diffusion Video Editing
Choose for a dedicated CLI/framework specializing in video upscaling and frame interpolation, without other AIGC tasks.
A machine learning-based video super resolution and frame interpolation framework. Est. Hack the Valley II, 2018.
Choose for a state-of-the-art large-scale video generation model when you don't need MMagic's unified zoo.
Wan: Open and Advanced Large-Scale Video Generative Models
Choose for the interactive image editing system from the CVPR paper over MMagic's general editing APIs.
[CVPR'25] Official Implementations for Paper - MagicQuill: An Intelligent Interactive Image Editing System
An earlier Wan version; select it if you specifically need that generation's checkpoints or codebase.
Wan: Open and Advanced Large-Scale Video Generative Models
Choose for a simple, proven structure-guided inpainting baseline without extra dependencies.
EdgeConnect: Structure Guided Image Inpainting using Edge Prediction, ICCV 2019 https://arxiv.org/abs/1901.00212
Does the same job, but ships as an app. Useful to a person, not swappable into a codebase.
A GUI app for video/image upscaling and frame interpolation; choose for desktop use instead of code integration.
Video, Image and GIF upscale/enlarge(Super-Resolution) and Video frame interpolation. Achieved with Waifu2x, Real-ESRGAN, Real-CUGAN, RTX Video Super Resolution VSR, SRMD, RealSR, Anime4K, RIFE, IFRNet, CAIN, DAIN, and ACNet.
A browser GUI for stable diffusion; pick it for interactive prompt-based image generation and editing, not as a library.
Stable Diffusion web UI
An end-user tool for interactive inpainting/object removal; choose for GUI-based photo editing rather than programmatic use.
Image inpainting tool powered by SOTA AI Model. Remove any unwanted object, defect, people from your pictures or erase and replace(powered by stable diffusion) any thing on your pictures.
An automated short-video engine/app, not a library for integrating generative AI into a codebase.
๐ AI ๅ จ่ชๅจ็ญ่ง้ขๅผๆ | AI Fully Automated Short Video Engine
These projects were analysed and named mmagic among their alternatives. The relationship is not symmetric โ how mmagic rates them is a separate judgement, made when mmagic is analysed in its own right.
calls mmagic โAn older OpenMMLab generative-AI toolbox with diffusion and GAN models; choose if you are already in the MM ecosystem, expecting stale maintenance.โ
๐ค Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch.