DragGAN is an interactive generative AI editor and GAN image editing tool designed for modifying the shape and structure of objects within images. It functions as a latent space manipulator that enables precise geometric and appearance editing by transforming images into editable latent codes. The system provides a web-based visual dashboard for real-time manipulation. Users can change the appearance of generated objects through an interactive point-based dragging interface, utilizing a process where source and target coordinates drive the optimization of the generative model. The project in
AnimeGANv2 is a generative adversarial network training framework and image stylization tool designed to convert real-world photographs and videos into anime-style imagery. It functions as an anime style generator that transforms real-world scenes into animation through supervised style transfer. The project provides a system for training style models and extracting specific generator weight parameters from deep learning checkpoints to create lightweight models for inference. It focuses on landscape image stylization and the ability to mimic specific artistic styles from provided datasets. T
This is a library of generative model architectures built using the TensorFlow framework. It provides implementations for producing synthetic data and realistic images, specifically focusing on Variational Autoencoders and various Generative Adversarial Network variants. The collection includes specific GAN architectures such as WGAN-GP, LSGAN, InfoGAN, and EBGAN. It also features Variational Autoencoders designed to learn latent representations and synthesize new samples from learned distributions. The project covers image processing pipelines for normalizing and cropping data, as well as a
mmagic is a multimodal training pipeline and framework for generative AI, focusing on visual synthesis and restoration. It provides the infrastructure to build and train models for tasks such as text-to-image and text-to-video generation, 3D-aware content synthesis, and high-fidelity image translation using diffusion models and generative adversarial networks. The project distinguishes itself through specialized capabilities for generative model personalization, including techniques for fine-tuning subjects and styles. It also supports advanced visual manipulations such as latent space interp
DragGAN est un éditeur et manipulateur d'images génératives interactif qui permet aux utilisateurs de remodeler la sortie visuelle en déplaçant des points de contrôle sur une variété de réseaux génératifs. Il fonctionne comme un outil d'édition d'image basée sur des points, mappant les déplacements de coordonnées définis par l'utilisateur vers l'espace latent d'un modèle génératif pour déformer les images.
Les fonctionnalités principales de opengvlab/draggan sont : Latent Space Image Manipulations, Latent Space Manipulations, Interactive GAN Editors, Interactive Manifold Editors, Point Tracking, GAN Inversion, Real Image Inversion, Iterative Image Optimizers.
Les alternatives open-source à opengvlab/draggan incluent : xingangpan/draggan — DragGAN is an interactive generative AI editor and GAN image editing tool designed for modifying the shape and… tachibanayoshino/animeganv2 — AnimeGANv2 is a generative adversarial network training framework and image stylization tool designed to convert… hwalsuklee/tensorflow-generative-model-collections — This is a library of generative model architectures built using the TensorFlow framework. It provides implementations… open-mmlab/mmagic — mmagic is a multimodal training pipeline and framework for generative AI, focusing on visual synthesis and… orpatashnik/styleclip — StyleCLIP is a framework for steering generative models using contrastive language-image pre-training to edit and… xinntao/esrgan — ESRGAN is a deep learning image restoration framework designed for image super-resolution. It uses a generative…