StudioGAN is a Pytorch library providing implementations of representative Generative Adversarial Networks (GANs) for conditional/unconditional image generation.

Official Implementation of "Third Time's the Charm? Image and Video Editing with StyleGAN3" (AIM ECCVW 2022) https://arxiv.org/abs/2201.13433

A collection of Jupyter notebooks to play with NVIDIA's StyleGAN3 and OpenAI's CLIP for a text-based guided image generation.

Unofficial implementation of DragGAN with StyleGAN2/3 pretrained models

InterfaceGAN : Exploring the limits of InterfaceGAN. In this repository, we propose an approach, termed as InterFaceGAN , for semantic face editing based on the work from Shen et al. Specifically, we leverage the ideas from the previous work, by applying the method for new face attributes, and also for StyleGAN3. We qualitatively explain that …

Efficient Scale-Invariant Generator with Column-Row Entangled Pixel Synthesis (CVPR 2023)

3DGANTex: 3D Face Reconstruction with StyleGAN3-based Texture Synthesis from Multi-View Images

Inverse an image to the latent space of StyleGAN3

AnimeGAN2 trained on Arcane

ncnn implementation of StyleGAN2ADA and StyleGAN3.

(2021) Robust Deepfake Detection project for the Deep Learning course at ETH. Authors: David Kamm, Nicolas Muntwyler, Alexander Timans, Moritz Vandenhirtz

uses OSC to modify stylegan3 generation and sends result via NDI. Comes with a touchdsigner project.

State-of-the-art https://arxiv.org/abs/2302.09119 https://intranet.matematicas.uady.mx/journal/index.php?c=50

audio reactive visuals with StyleGAN

Modifications of the official PyTorch implementation of StyleGAN3, using PDillis's fork as base and incorporating code from other repo's, eg OSMR's blending

The repository has scripts and notebooks to train generative models. We specifically aim to train histo-pathology images which are protected under HIPAA law, to make a robust dataset for future pathology computer vision endeavors.

This project investigates the integration of alias-free resampling techniques, inspired by StyleGAN3, into the UNet architecture of Diffusion models. Our approach focuses on improving model performance without introducing new trainable parameters, maintaining efficiency and simplicity.

A Gradio app for creating bros face pics using BroGAN

StyleGAN3 and other neural network models of the berlin based visionary artist Michael "Colory" Krebs.

Alias-Free Generative Adversarial Networks (StyleGAN3)