albertpumarola/GANimation
PyTorch implementation of a GAN that edits facial expressions in single images using Action Unit annotations to control expression magnitude.
Not currently ranked — collecting fresh signals.
star history
This repository contains the official implementation of a generative adversarial network for facial animation. The approach uses Action Units (AU) annotations to condition the GAN, enabling continuous control over the magnitude of facial muscle activations. The model can manipulate expressions by combining and adjusting individual Action Units, transforming a single input image into a sequence of animated facial expressions while preserving identity and lighting.
Frequently asked
- What is albertpumarola/GANimation?
- PyTorch implementation of a GAN that edits facial expressions in single images using Action Unit annotations to control expression magnitude.
- Is GANimation open source?
- Yes — albertpumarola/GANimation is open source, released under the GPL-3.0 license.
- What language is GANimation written in?
- albertpumarola/GANimation is primarily written in Python.
- How popular is GANimation?
- albertpumarola/GANimation has 2k stars on GitHub.
- Where can I find GANimation?
- albertpumarola/GANimation is on GitHub at https://github.com/albertpumarola/GANimation.