Tengfei-Wang/HFGI
A high-fidelity GAN inversion framework for editing image attributes while preserving details like background, appearance, and illumination.
Not currently ranked — collecting fresh signals.
star history
This repository implements a CVPR 2022 paper that introduces a novel approach to invert real images into the latent space of pretrained GAN models, enabling attribute editing with image-specific details preserved. The framework includes inference code, pretrained models, and training code. Users can run experiments via provided Jupyter notebooks or local environments.
Frequently asked
- What is Tengfei-Wang/HFGI?
- A high-fidelity GAN inversion framework for editing image attributes while preserving details like background, appearance, and illumination.
- Is HFGI open source?
- Yes — Tengfei-Wang/HFGI is an open-source project tracked on heatdrop.
- What language is HFGI written in?
- Tengfei-Wang/HFGI is primarily written in Jupyter Notebook.
- How popular is HFGI?
- Tengfei-Wang/HFGI has 476 stars on GitHub.
- Where can I find HFGI?
- Tengfei-Wang/HFGI is on GitHub at https://github.com/Tengfei-Wang/HFGI.