UnCRtainTS: Uncertainty Quantification for Cloud Removal in Optical Satellite Time Series
This work addresses the challenge of continuous Earth surface monitoring for remote sensing applications by providing explicit cloud removal with quality control, though it appears incremental as it builds on existing deep learning approaches.
The paper tackles the problem of cloud occlusion in optical satellite images by introducing UnCRtainTS, a method for multi-temporal cloud removal that combines an attention-based architecture with multivariate uncertainty prediction, achieving new state-of-the-art performance in image reconstruction on two public datasets.
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.