The competition aims at developing foundation deep learning models for full waveform inversion of multi-offset ground-penetrating radar (GPR) data. The models should be trained subject to synthetic data using gprMax. The participants can either use the provided synthetic datasets, or they can use gprMax to generate additional data to complement training. The participants can also apply any pre-processing they think necessary to the input files prior to training.
For more information check the Kaggle link https://www.kaggle.com/competitions/gpr-max-deep-learning-challenge-1-gdlc-1
Giannakis, I., Warren, C., Giannopoulos, A., Leontidis, G., Su, Y., Zhou, F., Martin-Torres, J., Diamanti, N., "Investigating the Capabilities of Deep Learning for Processing and Interpreting One-Shot Multi-offset GPR Data: A Numerical Case Study for Lunar and Martian Environments", arXiv:2410.14386, 2024