shepnerd/inpainting_gmcnn
A PyTorch/TensorFlow implementation of a GAN-based image inpainting model from NeurIPS 2018.
Not currently ranked — collecting fresh signals.
star history
This repository implements the GMCNN (Generative Multi-column Convolutional Neural Network) paper for filling missing regions in images. The model uses multiple CNN branches at different scales to reconstruct missing image content, trained with adversarial loss for realistic results. It supports both PyTorch and TensorFlow implementations and was evaluated on standard benchmarks including Places2, CelebA-HQ, and Paris streetview datasets.
Frequently asked
- What is shepnerd/inpainting_gmcnn?
- A PyTorch/TensorFlow implementation of a GAN-based image inpainting model from NeurIPS 2018.
- Is inpainting_gmcnn open source?
- Yes — shepnerd/inpainting_gmcnn is open source, released under the MIT license.
- What language is inpainting_gmcnn written in?
- shepnerd/inpainting_gmcnn is primarily written in Python.
- How popular is inpainting_gmcnn?
- shepnerd/inpainting_gmcnn has 435 stars on GitHub.
- Where can I find inpainting_gmcnn?
- shepnerd/inpainting_gmcnn is on GitHub at https://github.com/shepnerd/inpainting_gmcnn.