mlsanityData to 7 Oct 2026

Mission Critical -- Satellite Data is a Distinct Modality in Machine Learning

arXiv 2402.01444 ↗First posted 2 Feb 2024cs.LG cs.AI cs.CVcited by 76

Not on the map. Why?

Abstract (arXiv)

Satellite data has the potential to inspire a seismic shift for machine learning -- one in which we rethink existing practices designed for traditional data modalities. As machine learning for satellite data (SatML) gains traction for its real-world impact, our field is at a crossroads. We can either continue applying ill-suited approaches, or we can initiate a new research agenda that centers around the unique characteristics and challenges of satellite data. This position paper argues that satellite data constitutes a distinct modality for machine learning research and that we must recognize it as such to advance the quality and impact of SatML research across theory, methods, and deployment. We outline critical discussion questions and actionable suggestions to transform SatML from merely an intriguing application area to a dedicated research discipline that helps move the needle on big challenges for machine learning and society.

Key facts. Cited by 76 papers in the map (data through 7 Oct 2026). Builds on 62 of its 117 references that are in the map. First posted on arXiv 2 Feb 2024.

Builds on

62 of 117 references.

References · 62

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  4. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,131
  5. 2018Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentationcited by 3,210
  6. 2018Unsupervised Representation Learning by Predicting Image Rotationscited by 1,586
  7. 2018Model Cards for Model Reportingcited by 1,068
  8. 2020Measuring Robustness to Natural Distribution Shifts in Image Classificationcited by 452
  9. 2022SatMAE: Pre-training Transformers for Temporal and Multi-Spectral Satellite Imagerycited by 402
  10. 2017Functional Map of the Worldcited by 338
  11. 2022Scale-MAE: A Scale-Aware Masked Autoencoder for Multiscale Geospatial Representation Learningcited by 302
  12. 2019Tackling Climate Change with Machine Learningcited by 264
38 more
  1. 2019BigEarthNet: A Large-Scale Benchmark Archive For Remote Sensing Image Understandingcited by 231
  2. 2021Seasonal Contrast: Unsupervised Pre-Training from Uncurated Remote Sensing Datacited by 214
  3. 2020Geography-Aware Self-Supervised Learningcited by 190
  4. 2023GeoCLIP: Clip-Inspired Alignment between Locations and Images for Effective Worldwide Geo-localizationcited by 185
  5. 2023CROMA: Remote Sensing Representations with Contrastive Radar-Optical Masked Autoencoderscited by 167
  6. 2022SatlasPretrain: A Large-Scale Dataset for Remote Sensing Image Understandingcited by 164
  7. 2023SatCLIP: Global, General-Purpose Location Embeddings with Satellite Imagerycited by 148
  8. 2017Deep learning in remote sensing: a reviewcited by 128
  9. 2023GEO-Bench: Toward Foundation Models for Earth Monitoringcited by 122
  10. 2021Predicting with Confidence on Unseen Distributionscited by 116
  11. 2018Tile2Vec: Unsupervised representation learning for spatially distributed datacited by 109
  12. 2019Presence-Only Geographical Priors for Fine-Grained Image Classificationcited by 109
  13. 2022An Empirical Study of Remote Sensing Pretrainingcited by 105
  14. 2020Are Labels Always Necessary for Classifier Accuracy Evaluation?cited by 104
  15. 2019SEN12MS -- A Curated Dataset of Georeferenced Multi-Spectral Sentinel-1/2 Imagery for Deep Learning and Data Fusioncited by 95
  16. 2022Agreement-on-the-Line: Predicting the Performance of Neural Networks under Distribution Shiftcited by 94
  17. 2016Predicting Ground-Level Scene Layout from Aerial Imagerycited by 89
  18. 2020Multi-Scale Representation Learning for Spatial Feature Distributions using Grid Cellscited by 88
  19. 2020A Generalizable and Accessible Approach to Machine Learning with Global Satellite Imagerycited by 87
  20. 2022Deep Learning in Multimodal Remote Sensing Data Fusion: A Comprehensive Reviewcited by 73
  21. 2019Satellite Image Time Series Classification with Pixel-Set Encoders and Temporal Self-Attentioncited by 71
  22. 2021Recent Advances in Domain Adaptation for the Classification of Remote Sensing Datacited by 70
  23. 2023CSP: Self-Supervised Contrastive Spatial Pre-Training for Geospatial-Visual Representationscited by 67
  24. 2023Geographic Location Encoding with Spherical Harmonics and Sinusoidal Representation Networkscited by 61
  25. 2021SustainBench: Benchmarks for Monitoring the Sustainable Development Goals with Machine Learningcited by 53
  26. 2022Predicting Out-of-Distribution Error with the Projection Normcited by 50
  27. 2023SSL4EO-L: Datasets and Foundation Models for Landsat Imagerycited by 49
  28. 2019Building Damage Detection in Satellite Imagery Using Convolutional Neural Networkscited by 34
  29. 2018Land cover mapping at very high resolution with rotation equivariant CNNs: towards small yet accurate modelscited by 29
  30. 2018You Only Look Twice: Rapid Multi-Scale Object Detection In Satellite Imagerycited by 27
  31. 2021Towards a Collective Agenda on AI for Earth Science Data Analysiscited by 27
  32. 2023Revisiting pre-trained remote sensing model benchmarks: resizing and normalization matterscited by 27
  33. 2018Tiling and Stitching Segmentation Output for Remote Sensing: Basic Challenges and Recommendationscited by 24
  34. 2023Sat2Cap: Mapping Fine-Grained Textual Descriptions from Satellite Imagescited by 19
  35. 2023Better, Not Just More: Data-Centric Machine Learning for Earth Observationcited by 19
  36. 2020Estimating the Prediction Performance of Spatial Models via Spatial k-Fold Cross Validationcited by 14
  37. 2023Learning Tri-modal Embeddings for Zero-Shot Soundscape Mappingcited by 14
  38. 2020Assessing out-of-domain generalization for robust building damage detectioncited by 13

50 most cited of 62 All 62 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 22; 2025: 29; 2026: 25.

76 papers.

Sort:

By area:

  1. 2024OmniSat: Self-Supervised Modality Fusion for Earth Observationcited by 63
  2. 2024PANGAEA: A Global and Inclusive Benchmark for Geospatial Foundation Modelscited by 62
  3. 2025SegEarth-R1: Geospatial Pixel Reasoning via Large Language Modelcited by 40
  4. 2025Panopticon: Advancing Any-Sensor Foundation Models for Earth Observationcited by 34
  5. 2024GeoSynth: Contextually-Aware High-Resolution Satellite Image Synthesiscited by 32
  6. 2024On the Foundations of Earth and Climate Foundation Modelscited by 31
  7. 2024SegEarth-OV: Towards Training-Free Open-Vocabulary Segmentation for Remote Sensing Imagescited by 25
  8. 2024Specialized Foundation Models Struggle to Beat Supervised Baselinescited by 23
  9. 2024Mind the Modality Gap: Towards a Remote Sensing Vision-Language Model via Cross-modal Alignmentcited by 15
  10. 2025LISAT: Language-Instructed Segmentation Assistant for Satellite Imagerycited by 13
  11. 2026Foundation Models in Remote Sensing: Evolving from Unimodality to Multimodalitycited by 12
  12. 2024Distribution Shifts at Scale: Out-of-distribution Detection in Earth Observationcited by 10
38 more
  1. 2025SegEarth-R2: Towards Comprehensive Language-guided Segmentation for Remote Sensing Imagescited by 10
  2. 2025Geospatial Foundation Models to Enable Progress on Sustainable Development Goalscited by 9
  3. 2026No One Knows the State of the Art in Geospatial Foundation Modelscited by 9
  4. 2026EarthShift: a benchmark for measuring robustness to real-world distribution shifts in Earth observationcited by 8
  5. 2025Annotation-Free Open-Vocabulary Segmentation for Remote-Sensing Imagescited by 7
  6. 2024Contrasting local and global modeling with machine learning and satellite data: A case study estimating tree canopy height in African savannascited by 6
  7. 2025Be the Change You Want to See: Revisiting Remote Sensing Change Detection Practicescited by 6
  8. 2026THOR: A Versatile Foundation Model for Earth Observation Climate and Society Applicationscited by 6
  9. 2025How Does the Spatial Distribution of Pre-training Data Affect Geospatial Foundation Models?cited by 5
  10. 2025DescribeEarth: Describe Anything for Remote Sensing Imagescited by 5
  11. 2024Responsible AI for Earth Observationcited by 4
  12. 2025Thicker and Quicker: A Jumbo Token for Fast Plain Vision Transformerscited by 4
  13. 2025A Genealogy of Foundation Models in Remote Sensingcited by 4
  14. 2025GeoCrossBench: Cross-Band Generalization for Remote Sensingcited by 4
  15. 2024KidSat: satellite imagery to map childhood poverty dataset and benchmarkcited by 3
  16. 2024Prototype-Based Methods in Explainable AI and Emerging Opportunities in the Geosciencescited by 3
  17. 2025On the Effectiveness of Methods and Metrics for Explainable AI in Remote Sensing Image Scene Classificationcited by 3
  18. 2025Missing Data as Augmentation in the Earth Observation Domain: A Multi-View Learning Approachcited by 2
  19. 2025Time2Agri: Temporal Pretext Tasks for Agricultural Monitoringcited by 2
  20. 2025MAESTRO: Masked AutoEncoders for Multimodal, Multitemporal, and Multispectral Earth Observation Datacited by 2
  21. 2025Mapping on a Budget: Optimizing Spatial Data Collection for MLcited by 2
  22. 2025Multi-modal Co-learning for Earth Observation: Enhancing single-modality models via modality collaborationcited by 2
  23. 2025Mapping the Vanishing and Transformation of Urban Villages in Chinacited by 2
  24. 2025Are Neuro-Inspired Multi-Modal Vision-Language Models Resilient to Membership Inference Privacy Leakage?cited by 2
  25. 2026Brewing Stronger Features: Dual-Teacher Distillation for Multispectral Earth Observationcited by 2
  26. 2026Advancing Earth Observation Through Machine Learning: A TorchGeo Tutorialcited by 2
  27. 2026ConInfer: Context-Aware Inference for Training-Free Open-Vocabulary Remote Sensing Segmentationcited by 2
  28. 2026Are Pretrained Image Matchers Good Enough for SAR-Optical Satellite Registration?cited by 2
  29. 2024Impacts of Color and Texture Distortions on Earth Observation Data in Deep Learningcited by 1
  30. 2024Seeing Through the Clouds: Cloud Gap Imputation with Prithvi Foundation Modelcited by 1
  31. 2024SeeFar: Satellite Agnostic Multi-Resolution Dataset for Geospatial Foundation Modelscited by 1
  32. 2024Improving satellite imagery segmentation using multiple Sentinel-2 revisitscited by 1
  33. 2025Smooth Sailing: Lipschitz-Driven Uncertainty Quantification for Spatial Associationcited by 1
  34. 2025On the Generalization of Representation Uncertainty in Earth Observationcited by 1
  35. 2025GraphVSSM: Graph Variational State-Space Model for Probabilistic Spatiotemporal Inference of Dynamic Exposure and Vulnerability for Regional Disaster Resilience Assessmentcited by 1
  36. 2025A Spatially Masked Adaptive Gated Network for multimodal post-flood water extent mapping using SAR and incomplete multispectral datacited by 1
  37. 2026MOMO: Mars Orbital Model Foundation Model for Mars Orbital Applicationscited by 1
  38. 2026OT on the Map: Quantifying Domain Shifts in Geographic Spacecited by 1

50 most cited of 76 All 76 (JSON)