Diffusion Models for Interferometric Satellite Aperture Radar
This work addresses the need for synthetic labelled satellite data to support deep learning in interferometric satellite aperture radar processing, though it is incremental as it applies an existing method to a new domain.
The authors tackled the problem of generating synthetic radar-based satellite images using Probabilistic Diffusion Models (PDMs), showing that PDMs can produce complex and realistic structures but face issues with sampling time, as accelerated strategies fail on these datasets.
Probabilistic Diffusion Models (PDMs) have recently emerged as a very promising class of generative models, achieving high performance in natural image generation. However, their performance relative to non-natural images, like radar-based satellite data, remains largely unknown. Generating large amounts of synthetic (and especially labelled) satellite data is crucial to implement deep-learning approaches for the processing and analysis of (interferometric) satellite aperture radar data. Here, we leverage PDMs to generate several radar-based satellite image datasets. We show that PDMs succeed in generating images with complex and realistic structures, but that sampling time remains an issue. Indeed, accelerated sampling strategies, which work well on simple image datasets like MNIST, fail on our radar datasets. We provide a simple and versatile open-source https://github.com/thomaskerdreux/PDM_SAR_InSAR_generation to train, sample and evaluate PDMs using any dataset on a single GPU.