UnCRtainTS: Uncertainty Quantification for Cloud Removal in Optical Satellite Time Series

TL;DR

UnCRtainTS introduces an attention-based model with uncertainty prediction for improved cloud removal in satellite images, enabling better reconstruction quality control and setting new performance benchmarks.

Contribution

The paper presents a novel attention-based architecture combined with multivariate uncertainty estimation for multi-temporal cloud removal in satellite imagery.

Findings

Achieved state-of-the-art image reconstruction performance.

Well-calibrated uncertainties allow precise control of reconstruction quality.

Effective handling of diverse cloud occlusion scenarios.

Abstract

Clouds and haze often occlude optical satellite images, hindering continuous, dense monitoring of the Earth's surface. Although modern deep learning methods can implicitly learn to ignore such occlusions, explicit cloud removal as pre-processing enables manual interpretation and allows training models when only few annotations are available. Cloud removal is challenging due to the wide range of occlusion scenarios -- from scenes partially visible through haze, to completely opaque cloud coverage. Furthermore, integrating reconstructed images in downstream applications would greatly benefit from trustworthy quality assessment. In this paper, we introduce UnCRtainTS, a method for multi-temporal cloud removal combining a novel attention-based architecture, and a formulation for multivariate uncertainty prediction. These two components combined set a new state-of-the-art performance in terms of image reconstruction on two public cloud removal datasets. Additionally, we show how the well-calibrated predicted uncertainties enable a precise control of the reconstruction quality.