README.md

FWIGAN: Full-Waveform Inversion with Deep Adversarial Learning

This repo contains a PyTorch implementation with DeepWave for the FWIGAN based on the following papers:

1. Revisit Geophysical Imaging in A New View of Physics-informed Generative Adversarial Learning
2. CryoGAN: A New Reconstruction Paradigm for Single-Particle Cryo-EM Via Deep Adversarial Learning

I implemented the FWIGAN to invert the velocity model, source wavelet, and noise level in data based on the additive Gaussian noise. The observed data is generated by using densely discretized model to avoid inverse crimes in the inversion. Kindly check the Jupyter Notebooks for further details.