mlsanityData to 13 Oct 2026

SatlasPretrain: A Large-Scale Dataset for Remote Sensing Image Understanding

arXiv 2211.15660 ↗First posted 28 Nov 2022cs.CVcited by 165

Not on the map. Why?

Abstract (arXiv)

Remote sensing images are useful for a wide variety of planet monitoring applications, from tracking deforestation to tackling illegal fishing. The Earth is extremely diverse -- the amount of potential tasks in remote sensing images is massive, and the sizes of features range from several kilometers to just tens of centimeters. However, creating generalizable computer vision methods is a challenge in part due to the lack of a large-scale dataset that captures these diverse features for many tasks. In this paper, we present SatlasPretrain, a remote sensing dataset that is large in both breadth and scale, combining Sentinel-2 and NAIP images with 302M labels under 137 categories and seven label types. We evaluate eight baselines and a proposed method on SatlasPretrain, and find that there is substantial room for improvement in addressing research challenges specific to remote sensing, including processing image time series that consist of images from very different types of sensors, and taking advantage of long-range spatial context. Moreover, we find that pre-training on SatlasPretrain substantially improves performance on downstream tasks, increasing average accuracy by 18% over ImageNet and 6% over the next best baseline. The dataset, pre-trained model weights, and code are available at https://satlas-pretrain.allen.ai/.

Key facts. Cited by 165 papers in the map (data through 13 Oct 2026). Builds on 36 of its 72 references that are in the map. First posted on arXiv 28 Nov 2022.

Builds on

36 of 72 references.

References · 36

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,950
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,588
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,043
  4. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,111
  5. 2017Mask R-CNNcited by 7,266
  6. 2016Pyramid Scene Parsing Networkcited by 2,474
  7. 2021Masked-attention Mask Transformer for Universal Image Segmentationcited by 2,344
  8. 2020Improved Baselines with Momentum Contrastive Learningcited by 2,285
  9. 2017Rethinking Atrous Convolution for Semantic Image Segmentationcited by 2,159
  10. 2017Remote Sensing Image Scene Classification: Benchmark and State of the Artcited by 791
  11. 2017DOTA: A Large-scale Dataset for Object Detection in Aerial Imagescited by 622
  12. 2021Unifying Vision-and-Language Tasks via Text Generationcited by 411
24 more
  1. 2016AID: A Benchmark Dataset for Performance Evaluation of Aerial Scene Classificationcited by 344
  2. 2017Functional Map of the Worldcited by 338
  3. 2018DeepGlobe 2018: A Challenge to Parse the Earth through Satellite Imagescited by 277
  4. 2019BigEarthNet: A Large-Scale Benchmark Archive For Remote Sensing Image Understandingcited by 234
  5. 2021Seasonal Contrast: Unsupervised Pre-Training from Uncurated Remote Sensing Datacited by 215
  6. 2017LinkNet: Exploiting Encoder Representations for Efficient Semantic Segmentationcited by 212
  7. 2019iSAID: A Large-scale Dataset for Instance Segmentation in Aerial Imagescited by 196
  8. 2021UniT: Multimodal Multitask Learning with a Unified Transformercited by 195
  9. 2021Panoptic Segmentation of Satellite Image Time Series with Convolutional Temporal Attention Networkscited by 142
  10. 2018RoadTracer: Automatic Extraction of Road Networks from Aerial Imagescited by 102
  11. 2022SSL4EO-S12: A Large-Scale Multi-Modal, Multi-Temporal Dataset for Self-Supervised Learning in Earth Observationcited by 83
  12. 2018Topological Map Extraction from Overhead Imagescited by 58
  13. 2020Sat2Graph: Road Graph Extraction through Graph-Tensor Encodingcited by 55
  14. 2021ISTR: End-to-End Instance Segmentation with Transformerscited by 52
  15. 2021Self-Supervised Material and Texture Representation Learning for Remote Sensing Taskscited by 49
  16. 2021Multi-Modal Temporal Attention Models for Crop Mapping from Satellite Time Seriescited by 49
  17. 2021Towards General Purpose Vision Systemscited by 46
  18. 2022Webly Supervised Concept Expansion for General Purpose Vision Modelscited by 45
  19. 2021PolyWorld: Polygonal Building Extraction with Graph Neural Networks in Satellite Imagescited by 43
  20. 2022xView3-SAR: Detecting Dark Fishing Activity Using Synthetic Aperture Radar Imagerycited by 25
  21. 2019Human-Machine Collaboration for Fast Land Cover Mappingcited by 9
  22. 2021Beyond Road Extraction: A Dataset for Map Update using Aerial Imagescited by 5
  23. 2022A General Purpose Neural Architecture for Geospatial Systemscited by 5
  24. 2021Updating Street Maps using Changes Detected in Satellite Imagerycited by 3

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2023: 8; 2024: 45; 2025: 64; 2026: 48.

165 papers.

Sort:

By area:

  1. 2023SkySense: A Multi-Modal Remote Sensing Foundation Model Towards Universal Interpretation for Earth Observation Imagerycited by 203
  2. 2024Prithvi-EO-2.0: A Versatile Multi-Temporal Foundation Model for Earth Observation Applicationscited by 143
  3. 2024Neural Plasticity-Inspired Multimodal Foundation Model for Earth Observationcited by 102
  4. 2023Vision-Language Models in Remote Sensing: Current Progress and Future Trendscited by 93
  5. 2024EarthDial: Turning Multi-sensory Earth Observations to Interactive Dialoguescited by 81
  6. 2024Mission Critical -- Satellite Data is a Distinct Modality in Machine Learningcited by 77
  7. 2024MMEarth: Exploring Multi-Modal Pretext Tasks For Geospatial Representation Learningcited by 77
  8. 2024AnySat: One Earth Observation Model for Many Resolutions, Scales, and Modalitiescited by 73
  9. 2024HyperSIGMA: Hyperspectral Intelligence Comprehension Foundation Modelcited by 72
  10. 2024MTP: Advancing Remote Sensing Foundation Model via Multi-Task Pretrainingcited by 66
  11. 2024OmniSat: Self-Supervised Modality Fusion for Earth Observationcited by 63
  12. 2025Towards a Unified Copernicus Foundation Model for Earth Visioncited by 53
38 more
  1. 2024Locate Anything on Earth: Advancing Open-Vocabulary Object Detection for Remote Sensing Communitycited by 50
  2. 2025Galileo: Learning Global & Local Features of Many Remote Sensing Modalitiescited by 42
  3. 2024MetaEarth: A Generative Foundation Model for Global-Scale Remote Sensing Image Generationcited by 41
  4. 2024Changen2: Multi-Temporal Remote Sensing Generative Change Foundation Modelcited by 37
  5. 2025TESSERA: Temporal Embeddings of Surface Spectra for Earth Representation and Analysiscited by 36
  6. 2024One for All: Toward Unified Foundation Models for Earth Visioncited by 34
  7. 2025Panopticon: Advancing Any-Sensor Foundation Models for Earth Observationcited by 34
  8. 2024Major TOM: Expandable Datasets for Earth Observationcited by 31
  9. 2024On the Foundations of Earth and Climate Foundation Modelscited by 31
  10. 2024GeoLLM-Engine: A Realistic Environment for Building Geospatial Copilotscited by 28
  11. 2024Multi-Label Guided Soft Contrastive Learning for Efficient Earth Observation Pretrainingcited by 28
  12. 2025SkySense V2: A Unified Foundation Model for Multi-modal Remote Sensingcited by 26
  13. 2025TerraTorch: The Geospatial Foundation Models Toolkitcited by 25
  14. 2025TerraFM: A Scalable Foundation Model for Unified Multisensor Earth Observationcited by 25
  15. 2024Specialized Foundation Models Struggle to Beat Supervised Baselinescited by 24
  16. 2024ChatEarthNet: A Global-Scale Image-Text Dataset Empowering Vision-Language Geo-Foundation Modelscited by 23
  17. 2025OlmoEarth: Stable Latent Image Modeling for Multimodal Earth Observationcited by 21
  18. 2023Better, Not Just More: Data-Centric Machine Learning for Earth Observationcited by 19
  19. 2025A Survey on Remote Sensing Foundation Models: From Vision to Multimodalitycited by 19
  20. 2025Vision-Language Modeling Meets Remote Sensing: Models, Datasets and Perspectivescited by 18
  21. 2024Open-Canopy: Towards Very High Resolution Forest Monitoringcited by 16
  22. 2025RingMoE: Mixture-of-Modality-Experts Multi-Modal Foundation Models for Universal Remote Sensing Image Interpretationcited by 16
  23. 2024RingMo-Aerial: An Aerial Remote Sensing Foundation Model With Affine Transformation Contrastive Learningcited by 15
  24. 2025ROS-SAM: High-Quality Interactive Segmentation for Remote Sensing Moving Objectcited by 15
  25. 2025TerraMesh: A Planetary Mosaic of Multimodal Earth Observation Datacited by 14
  26. 2025REOBench: Benchmarking Robustness of Earth Observation Foundation Modelscited by 14
  27. 2024Low-Resource Vision Challenges for Foundation Modelscited by 13
  28. 2024WildSAT: Learning Satellite Image Representations from Wildlife Observationscited by 13
  29. 2024Harnessing Massive Satellite Imagery with Efficient Masked Image Modelingcited by 12
  30. 2024Global and Dense Embeddings of Earth: Major TOM Floating in the Latent Spacecited by 12
  31. 2025ATRNet-STAR: A Large Dataset and Benchmark Towards Remote Sensing Object Recognition in the Wildcited by 12
  32. 2026Foundation Models in Remote Sensing: Evolving from Unimodality to Multimodalitycited by 12
  33. 2026No One Knows the State of the Art in Geospatial Foundation Modelscited by 12
  34. 2024A2-MAE: A spatial-temporal-spectral unified remote sensing pre-training method based on anchor-aware masked autoencodercited by 11
  35. 2024RS-vHeat: Heat Conduction Guided Efficient Remote Sensing Foundation Modelcited by 11
  36. 2024SeaMo: A Season-Aware Multimodal Foundation Model for Remote Sensingcited by 11
  37. 2024Contrastive ground-level image and remote sensing pre-training improves representation learning for natural world imagerycited by 10
  38. 2025Can Generative Geospatial Diffusion Models Excel as Discriminative Geospatial Foundation Models?cited by 10

50 most cited of 165 All 165 (JSON)