yerfor/GeneFace
GeneFace is a neural network system that synthesizes high-fidelity 3D talking face videos from audio input using NeRF and PyTorch.
Not currently ranked — collecting fresh signals.
star history
GeneFace is an ICLR 2023 paper implementation that generates realistic talking face videos from audio. It uses neural radiance fields (NeRF) architecture with deep learning to achieve lip synchronization and facial expressiveness from speech input. The system handles out-of-domain audio generalization and includes follow-up works MimicTalk for person-specific rendering and GeneFace++ for improved quality.
Frequently asked
- What is yerfor/GeneFace?
- GeneFace is a neural network system that synthesizes high-fidelity 3D talking face videos from audio input using NeRF and PyTorch.
- Is GeneFace open source?
- Yes — yerfor/GeneFace is open source, released under the MIT license.
- What language is GeneFace written in?
- yerfor/GeneFace is primarily written in Python.
- How popular is GeneFace?
- yerfor/GeneFace has 2.7k stars on GitHub.
- Where can I find GeneFace?
- yerfor/GeneFace is on GitHub at https://github.com/yerfor/GeneFace.