mlsanityData to 7 Oct 2026

Self-supervised training for high-resolution close-range multispectral remote sensing imagery

arXiv 2607.11366 ↗First posted 13 Jul 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Although self-supervised learning (SSL) offers a promising way to reduce annotation effort in close-range remote sensing, its effectiveness for high-resolution multispectral unmanned aerial vehicle (UAV) imagery remains underexplored due to limited data. This study evaluated SSL pretraining for precision agriculture using cm-scale multispectral drone imagery collected across multiple sensors, years, and regions. Transformer-based encoders were pretrained with Momentum Contrast v3 (MoCo-v3) and Masked Autoencoders on a harmonized dataset combining msuav500K with newly collected multi-year UAV imagery from agricultural fields in Finland. Pretraining used four spectral bands (Green, Red, Red-Edge, Near-Infrared) for cross-sensor compatibility. The models were evaluated on crop-weed semantic segmentation using the WeedMap dataset with 5--100% training data. The following two subsets served as downstream tasks: Task A (Germany, RedEdge-M), where all pretrained models were compared under partial and full fine-tuning, and Task B (Switzerland, Sequoia), where the best encoder from Task A was assessed. Our Swin Transformer pretrained with MoCo-v3 achieved the strongest performance on both tasks, surpassing the Swin Transformer model of Doornbos et al. pretrained on a pre-release of msuav500K. Our pretrained Swin Transformer further demonstrated cross-sensor and cross-region generalization. We additionally provide a public multi-year multispectral UAV dataset from Finland to support future research.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 32 of its 46 references that are in the map. First posted on arXiv 13 Jul 2026.

Builds on

32 of 46 references.

References · 32

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  4. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  5. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
  6. 2023Segment Anythingcited by 8,158
  7. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,764
  8. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,482
  9. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,339
  10. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
  11. 2023Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detectioncited by 3,274
  12. 2020Improved Baselines with Momentum Contrastive Learningcited by 2,282
20 more
  1. 2021Vision Transformers for Dense Predictioncited by 1,789
  2. 2017EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classificationcited by 1,488
  3. 2025DINOv3cited by 1,297
  4. 2022Scale-MAE: A Scale-Aware Masked Autoencoder for Multiscale Geospatial Representation Learningcited by 302
  5. 2020Geography-Aware Self-Supervised Learningcited by 190
  6. 2015On Wasserstein Two Sample Testing and Related Families of Nonparametric Testscited by 165
  7. 2024Prithvi-EO-2.0: A Versatile Multi-Temporal Foundation Model for Earth Observation Applicationscited by 139
  8. 2025TerraMind: Large-Scale Generative Multimodality for Earth Observationcited by 101
  9. 2022SSL4EO-S12: A Large-Scale Multi-Modal, Multi-Temporal Dataset for Self-Supervised Learning in Earth Observationcited by 82
  10. 2018DeepWeeds: A Multiclass Weed Species Image Dataset for Deep Learningcited by 69
  11. 2024Vision Foundation Models in Remote Sensing: A Surveycited by 52
  12. 2025Towards a Unified Copernicus Foundation Model for Earth Visioncited by 52
  13. 2023SSL4EO-L: Datasets and Foundation Models for Landsat Imagerycited by 49
  14. 2024SpectralEarth: Training Hyperspectral Foundation Models at Scalecited by 42
  15. 2023PhenoBench -- A Large Dataset and Benchmarks for Semantic Image Interpretation in the Agricultural Domaincited by 31
  16. 2021A Survey of Deep Learning Techniques for Weed Detection from Imagescited by 23
  17. 2018WeedMap: A large-scale semantic weed mapping framework using aerial multispectral imaging and deep neural network for precision farmingcited by 22
  18. 2023Revisiting Evaluation Metrics for Semantic Segmentation: Optimization and Evaluation of Fine-grained Intersection over Unioncited by 17
  19. 2025Few-Shot Adaptation of Grounding DINO for Agricultural Domaincited by 5
  20. 2025WeedNet: A Foundation Model-Based Global-to-Local AI Approach for Real-Time Weed Species Identification and Classificationcited by 4

Built on by

Not cited yet.