mlsanityData to 7 Oct 2026

Masked Angle-Aware Autoencoder for Remote Sensing Images

arXiv 2408.01946 ↗First posted 4 Aug 2024cs.CVcited by 22

Not on the map. Why?

Abstract (arXiv)

To overcome the inherent domain gap between remote sensing (RS) images and natural images, some self-supervised representation learning methods have made promising progress. However, they have overlooked the diverse angles present in RS objects. This paper proposes the Masked Angle-Aware Autoencoder (MA3E) to perceive and learn angles during pre-training. We design a textitscaling center crop operation to create the rotated crop with random orientation on each original image, introducing the explicit angle variation. MA3E inputs this composite image while reconstruct the original image, aiming to effectively learn rotation-invariant representations by restoring the angle variation introduced on the rotated crop. To avoid biases caused by directly reconstructing the rotated crop, we propose an Optimal Transport (OT) loss that automatically assigns similar original image patches to each rotated crop patch for reconstruction. MA3E demonstrates more competitive performance than existing pre-training methods on seven different RS image datasets in three downstream tasks.

Key facts. Cited by 22 papers in the map (data through 7 Oct 2026). Builds on 54 of its 72 references that are in the map. First posted on arXiv 4 Aug 2024.

Builds on

54 of 72 references.

References · 54

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  3. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
  4. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  5. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,482
  6. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,339
  7. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
  8. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,451
  9. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
  10. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 4,993
  11. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,712
  12. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,875
38 more
  1. 2021BEiT: BERT Pre-Training of Image Transformerscited by 2,360
  2. 2013Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distancescited by 2,277
  3. 2019RandAugment: Practical automated data augmentation with a reduced search spacecited by 2,261
  4. 2021Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutionscited by 1,960
  5. 2021An Empirical Study of Training Self-Supervised Vision Transformerscited by 1,610
  6. 2017Accurate, Large Minibatch SGD: Training ImageNet in 1 Hourcited by 1,479
  7. 2018Unified Perceptual Parsing for Scene Understandingcited by 1,265
  8. 2021SimMIM: A Simple Framework for Masked Image Modelingcited by 1,147
  9. 2016Semantic Understanding of Scenes through the ADE20K Datasetcited by 1,063
  10. 2016Deep Networks with Stochastic Depthcited by 973
  11. 2017Remote Sensing Image Scene Classification: Benchmark and State of the Artcited by 791
  12. 2022Exploring Plain Vision Transformer Backbones for Object Detectioncited by 723
  13. 2017DOTA: A Large-scale Dataset for Object Detection in Aerial Imagescited by 622
  14. 2021Masked Feature Prediction for Self-Supervised Visual Pre-Trainingcited by 573
  15. 2017Large Batch Training of Convolutional Networkscited by 529
  16. 2022SatMAE: Pre-training Transformers for Temporal and Multi-Spectral Satellite Imagerycited by 402
  17. 2016AID: A Benchmark Dataset for Performance Evaluation of Aerial Scene Classificationcited by 343
  18. 2017Functional Map of the Worldcited by 338
  19. 2022Scale-MAE: A Scale-Aware Masked Autoencoder for Multiscale Geospatial Representation Learningcited by 302
  20. 2019Tackling Climate Change with Machine Learningcited by 264
  21. 2021Oriented R-CNN for Object Detectioncited by 230
  22. 2021Seasonal Contrast: Unsupervised Pre-Training from Uncurated Remote Sensing Datacited by 214
  23. 2019iSAID: A Large-scale Dataset for Instance Segmentation in Aerial Imagescited by 195
  24. 2021OTA: Optimal Transport Assignment for Object Detectioncited by 177
  25. 2022Advancing Plain Vision Transformer Towards Remote Sensing Foundation Modelcited by 149
  26. 2016Discrete Variational Autoencoderscited by 124
  27. 2022ViTAEv2: Vision Transformer Advanced by Exploring Inductive Bias for Image Recognition and Beyondcited by 117
  28. 2023Towards Geospatial Foundation Models via Continual Pretrainingcited by 109
  29. 2022ConvMAE: Masked Convolution Meets Masked Autoencoderscited by 105
  30. 2021Anchor-free Oriented Proposal Generator for Object Detectioncited by 104
  31. 2022Masked Frequency Modeling for Self-Supervised Visual Pre-Trainingcited by 70
  32. 2022MILAN: Masked Image Pretraining on Language Assisted Representationcited by 69
  33. 2022Uniform Masking: Enabling MAE Pre-training for Pyramid-based Vision Transformers with Localitycited by 57
  34. 2022Green Hierarchical Vision Transformer for Masked Image Modelingcited by 50
  35. 2023CMID: A Unified Self-Supervised Learning Framework for Remote Sensing Image Understandingcited by 48
  36. 2022MixMAE: Mixed and Masked Autoencoder for Efficient Pretraining of Hierarchical Vision Transformerscited by 46
  37. 2023Masked Image Modeling with Local Multi-Scale Reconstructioncited by 43
  38. 2023PixMIM: Rethinking Pixel Reconstruction in Masked Image Modelingcited by 28

50 most cited of 54 All 54 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 2; 2025: 14; 2026: 6.

22 papers.

Sort:

By area:

  1. 2024AnySat: One Earth Observation Model for Many Resolutions, Scales, and Modalitiescited by 73
  2. 2025Towards a Unified Copernicus Foundation Model for Earth Visioncited by 52
  3. 2025RemoteSAM: Towards Segment Anything for Earth Observationcited by 38
  4. 2025TESSERA: Temporal Embeddings of Surface Spectra for Earth Representation and Analysiscited by 35
  5. 2025SkySense V2: A Unified Foundation Model for Multi-modal Remote Sensingcited by 26
  6. 2025A Survey on Remote Sensing Foundation Models: From Vision to Multimodalitycited by 18
  7. 2026Foundation Models in Remote Sensing: Evolving from Unimodality to Multimodalitycited by 12
  8. 2025RoMA: Scaling up Mamba-based Foundation Models for Remote Sensingcited by 11
  9. 2026No One Knows the State of the Art in Geospatial Foundation Modelscited by 9
  10. 2025Visual Instruction Pretraining for Domain-Specific Foundation Modelscited by 7
  11. 2026RemoteAgent: Bridging Vague Human Intents and Earth Observation with RL-based Agentic MLLMscited by 7
  12. 2025TiMo: Spatiotemporal Foundation Model for Satellite Image Time Seriescited by 5
10 more
  1. 2025REJEPA: A Novel Joint-Embedding Predictive Architecture for Efficient Remote Sensing Image Retrievalcited by 4
  2. 2025A Genealogy of Foundation Models in Remote Sensingcited by 4
  3. 2025S5: Scalable Semi-Supervised Semantic Segmentation in Remote Sensingcited by 3
  4. 2024Detecting Looted Archaeological Sites from Satellite Image Time Seriescited by 2
  5. 2025MAESTRO: Masked AutoEncoders for Multimodal, Multitemporal, and Multispectral Earth Observation Datacited by 2
  6. 2026Brewing Stronger Features: Dual-Teacher Distillation for Multispectral Earth Observationcited by 2
  7. 2026Pretrain Where? Investigating How Pretraining Data Diversity Impacts Geospatial Foundation Model Performancecited by 1
  8. 2025REMSA: Foundation Model Selection for Remote Sensing via a Constraint-Aware Agentcited by 0
  9. 2025MMLANDMARKS: a Cross-View Instance-Level Benchmark for Geo-Spatial Understandingcited by 0
  10. 2026Spectrally Distilled Representations Aligned with Instruction-Augmented LLMs for Satellite Imagerycited by 0