HighRes-net: Recursive Fusion for Multi-Frame Super-Resolution of Satellite Imagery

TL;DR

HighRes-net is an end-to-end deep learning model for multi-frame satellite image super-resolution that learns co-registration, fusion, and up-sampling, achieving state-of-the-art results in real-world applications.

Contribution

It introduces the first deep learning approach to MFSR that learns all sub-tasks jointly and recursively, including implicit co-registration and a novel registered loss.

Findings

Topped the ESA MFSR competition with real satellite data.

Effectively learns co-registration without explicit mechanisms.

Enhances Earth Observation imagery at scale.

Abstract

Generative deep learning has sparked a new wave of Super-Resolution (SR) algorithms that enhance single images with impressive aesthetic results, albeit with imaginary details. Multi-frame Super-Resolution (MFSR) offers a more grounded approach to the ill-posed problem, by conditioning on multiple low-resolution views. This is important for satellite monitoring of human impact on the planet -- from deforestation, to human rights violations -- that depend on reliable imagery. To this end, we present HighRes-net, the first deep learning approach to MFSR that learns its sub-tasks in an end-to-end fashion: (i) co-registration, (ii) fusion, (iii) up-sampling, and (iv) registration-at-the-loss. Co-registration of low-resolution views is learned implicitly through a reference-frame channel, with no explicit registration mechanism. We learn a global fusion operator that is applied recursively on an arbitrary number of low-resolution pairs. We introduce a registered loss, by learning to align the SR output to a ground-truth through ShiftNet. We show that by learning deep representations of multiple views, we can super-resolve low-resolution signals and enhance Earth Observation data at scale. Our approach recently topped the European Space Agency's MFSR competition on real-world satellite imagery.