Diffusion Models for Interferometric Satellite Aperture Radar

TL;DR

This paper explores the application of Probabilistic Diffusion Models to generate synthetic interferometric satellite radar images, demonstrating their potential and challenges in handling complex radar data for deep learning tasks.

Contribution

It introduces the use of PDMs for radar satellite data generation and evaluates their effectiveness and sampling issues, providing an open-source tool for the community.

Findings

PDMs can generate realistic radar satellite images with complex structures.

Sampling time remains a challenge for radar datasets.

Accelerated sampling strategies are less effective on radar data.

Abstract

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.