Toward Learning Latent-Variable Representations of Microstructures by Optimizing in Spatial Statistics Space

no code implementations • 16 Feb 2024 • Sayed Sajad Hashemi, Michael Guerzhoy, Noah H. Paulson

In this work, we train a Variational Autoencoders (VAE) to produce reconstructions of textures that preserve the spatial statistics of the original texture, while not necessarily reconstructing the same image in data space.