mlsanityData to 14 Oct 2026

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

arXiv 2304.05464 ↗First posted 11 Apr 2023cs.CV eess.IVcited by 28

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 28 papers in the map (data through 14 Oct 2026). Builds on 41 of its 77 references that are in the map. First posted on arXiv 11 Apr 2023.

Builds on

41 of 77 references.

References · 41

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  3. 2017On Calibration of Modern Neural Networkscited by 4,960
  4. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,795
  5. 2017Squeeze-and-Excitation Networkscited by 4,789
  6. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,527
  7. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,448
  8. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,926
  9. 2016Perceptual Losses for Real-Time Style Transfer and Super-Resolutioncited by 2,694
  10. 2017What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?cited by 2,352
  11. 2014Image Super-Resolution Using Deep Convolutional Networkscited by 1,224
  12. 2017Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networkscited by 707
29 more
  1. 2021A Survey of Uncertainty in Deep Neural Networkscited by 680
  2. 2021Panoptic Segmentation of Satellite Image Time Series with Convolutional Temporal Attention Networkscited by 143
  3. 2022On the Pitfalls of Heteroscedastic Uncertainty Estimation with Probabilistic Neural Networkscited by 120
  4. 2022A high-resolution canopy height model of the Earthcited by 91
  5. 2019A Deep Journey into Super-resolution: A surveycited by 86
  6. 2019Self-attention for raw optical Satellite Time Series Classificationcited by 75
  7. 2019Satellite Image Time Series Classification with Pixel-Set Encoders and Temporal Self-Attentioncited by 71
  8. 2020On hallucinations in tomographic image reconstructioncited by 68
  9. 2022MR Image Denoising and Super-Resolution Using Regularized Reverse Diffusioncited by 67
  10. 2019Reliable training and estimation of variance networkscited by 63
  11. 2020Lightweight Temporal Self-Attention for Classifying Satellite Image Time Seriescited by 55
  12. 2019Uncertainty Quantification in Deep MRI Reconstructioncited by 49
  13. 2018Super-resolution of Sentinel-2 images: Learning a globally applicable deep neural networkcited by 44
  14. 2021Crop mapping from image time series: deep learning with multi-scale label hierarchiescited by 42
  15. 2022Faithful Heteroscedastic Regression with Neural Networkscited by 37
  16. 2020Multi-Sensor Data Fusion for Cloud Removal in Global and All-Season Sentinel-2 Imagerycited by 36
  17. 2022SEN12MS-CR-TS: A Remote Sensing Data Set for Multi-modal Multi-temporal Cloud Removalcited by 32
  18. 2020Cloud Removal for Remote Sensing Imagery via Spatial Attention Generative Adversarial Networkcited by 29
  19. 2017Filmy Cloud Removal on Satellite Imagery with Multispectral Conditional Generative Adversarial Netscited by 26
  20. 2021Global canopy height regression and uncertainty estimation from GEDI LIDAR waveforms with deep ensemblescited by 26
  21. 2022FiLM-Ensemble: Probabilistic Deep Learning via Feature-wise Linear Modulationcited by 24
  22. 2022GLF-CR: SAR-Enhanced Cloud Removal with Global-Local Fusioncited by 24
  23. 2020Uncertainty Estimation in Medical Image Denoising with Bayesian Deep Image Priorcited by 18
  24. 2022Autoinverse: Uncertainty Aware Inversion of Neural Networkscited by 16
  25. 2020Crop Classification under Varying Cloud Cover with Neural Ordinary Differential Equationscited by 12
  26. 2020Cloud removal in remote sensing images using generative adversarial networks and SAR-to-optical image translationcited by 11
  27. 2022Uncertainty Estimation for Computed Tomography with a Linearised Deep Image Priorcited by 10
  28. 2022Generalized sparse Bayesian learning and application to image reconstructioncited by 6
  29. 2022Estimating Regression Predictive Distributions with Sample Networkscited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 3; 2024: 7; 2025: 8; 2026: 10.

28 papers.

Sort:

By area:

  1. 2023DiffCR: A Fast Conditional Diffusion Framework for Cloud Removal from Optical Satellite Imagescited by 27
  2. 2024Diffusion Models Meet Remote Sensing: Principles, Methods, and Perspectivescited by 27
  3. 2024IDF-CR: Iterative Diffusion Process for Divide-and-Conquer Cloud Removal in Remote-sensing Imagescited by 13
  4. 2025Effective Cloud Removal for Remote Sensing Images by an Improved Mean-Reverting Denoising Model with Elucidated Design Spacecited by 12
  5. 2024AllClear: A Comprehensive Dataset and Benchmark for Cloud Removal in Satellite Imagerycited by 7
  6. 2025FLAIR-HUB: Large-scale Multimodal Dataset for Land Cover and Crop Mappingcited by 6
  7. 2024Attentive Contextual Attention for Cloud Removalcited by 5
  8. 2023Diffusion Models for Earth Observation Use-cases: from cloud removal to urban change detectioncited by 4
  9. 2025Missing Data as Augmentation in the Earth Observation Domain: A Multi-View Learning Approachcited by 2
  10. 2025Multimodal Diffusion Bridge with Attention-Based SAR Fusion for Satellite Image Cloud Removalcited by 2
  11. 2025An Analysis of Temporal Dropout in Earth Observation Time Series for Regression Taskscited by 2
  12. 2025UniTS: Unified Spatio-Temporal Generative Model for Remote Sensingcited by 2
16 more
  1. 2024In the Search for Optimal Multi-view Learning Models for Crop Classification with Global Remote Sensing Datacited by 1
  2. 2025SAR-to-RGB Translation with Latent Diffusion for Earth Observationcited by 1
  3. 2025Cloud-Aware SAR Fusion for Enhanced Optical Sensing in Space Missionscited by 1
  4. 2026SADER: Structure-Aware Diffusion Framework with DEterministic Resampling for Multi-Temporal Remote Sensing Cloud Removalcited by 1
  5. 2026CBEN -- A Multimodal Machine Learning Dataset for Cloud Robust Remote Sensing Image Understandingcited by 1
  6. 2026Asynchronous Remote Sensing Time-Series Fusion for Cloud Removal and Anytime Reconstructioncited by 1
  7. 2026ATT-CR: Adaptive Triangular Transformer for Cloud Removalcited by 1
  8. 2026Difference-Driven Gating: Adaptive Feature Fusion for U-Net Decodercited by 1
  9. 2023Quantum Machine Learning for Remote Sensing: Exploring potential and challengescited by 0
  10. 2024Ai4Fapar: How artificial intelligence can help to forecast the seasonal earth observation signalcited by 0
  11. 2024Leveraging Multi-Temporal Sentinel 1 and 2 Satellite Data for Leaf Area Index Estimation With Deep Learningcited by 0
  12. 2026Self-Supervised Uncertainty Estimation For Super-Resolution of Satellite Imagescited by 0
  13. 2026IB-HFN: Information Bottleneck-Driven SAR-Optical Fusion Network for High-Fidelity Cloud Removalcited by 0
  14. 2026From Surface Forecasting to Observability Forecasting: A Latent World Model for Cloud-Aware EO Monitoringcited by 0
  15. 2026GeoCore-9B: Towards Geo-Aware Generative Foundation Models in Earth Observationcited by 0
  16. 2026GeoCR: Learning a Generalist Cloud Removal Prior from Heterogeneous Observationscited by 0