microsoft/Deep3DFaceReconstruction
A TensorFlow implementation of CNN-based 3D face reconstruction from single images using weakly-supervised learning.
Not currently ranked — collecting fresh signals.
star history
This repository provides a deep learning method for accurate 3D face reconstruction from 2D images. It uses a hybrid-level weakly-supervised training approach with CNNs to reconstruct face geometry and high-fidelity textures from single photos, handling pose variations and occlusions. The project implements the 2019 CVPRW paper that won the Best Paper Award and has been superseded by a higher-performing PyTorch implementation.
Frequently asked
- What is microsoft/Deep3DFaceReconstruction?
- A TensorFlow implementation of CNN-based 3D face reconstruction from single images using weakly-supervised learning.
- Is Deep3DFaceReconstruction open source?
- Yes — microsoft/Deep3DFaceReconstruction is open source, released under the MIT license.
- What language is Deep3DFaceReconstruction written in?
- microsoft/Deep3DFaceReconstruction is primarily written in Python.
- How popular is Deep3DFaceReconstruction?
- microsoft/Deep3DFaceReconstruction has 2.3k stars on GitHub.
- Where can I find Deep3DFaceReconstruction?
- microsoft/Deep3DFaceReconstruction is on GitHub at https://github.com/microsoft/Deep3DFaceReconstruction.