Clustering Guided Domain-Specific Pretrained Foundation Model Very High-Resolution Arctic Remote Sensing
Quick Take
This study presents a novel Arctic-focused remote sensing foundation model (RSFM) utilizing a Vision Transformer (ViT) encoder pretrained with domain-specific data, achieving significant improvements in F1 scores across four datasets. The model outperformed existing benchmarks, demonstrating that tailored self-supervised pretraining enhances fine-scale Arctic mapping capabilities.
Key Points
- Developed Arctic-focused RSFM using 3 million chips from 267 TB of VHSR imagery.
- ViT-Large encoder pretrained with domain-adapted MAE achieved F1 score improvements of 5-8%.
- Outperformed Prithvi-EO-2.0 in all downstream tasks with at least 15% F1 score gain.
- Optimized pretraining data distribution enhances Arctic mapping representation transferability.
- Model integrated into existing detection and segmentation frameworks for practical applications.
Article Content
From source RSS / original summaryarXiv:2605. 30467v1 Announce Type: new Abstract: This study introduces a novel Arctic-focused remote sensing foundation model (RSFM) by combining diversity-aware regional-scale image curation with masked autoencoder (MAE) self-supervised pretraining of a Vision Transformer (ViT) encoder for very-high-spatial-resolution (VHSR) satellite image analysis.
Spectral and acquisition-metadata descriptors were used in a scalable affinity-propagation clustering workflow to select approximately 3 million chips from 267 TB of Vantor VHSR imagery This curation strategy was designed to reduce oversampling of visually repetitive or low-information areas while preserving broad scene diversity across the study domain.
We pretrained a ViT-Large encoder on the curated corpus using a domain-adapted MAE reconstruction objective, producing Arctic-specific transformer weights for downstream feature mapping. The pretrained encoder was integrated into an existing location-aware detection and segmentation framework and evaluated across four hand-labeled Arctic datasets. Compared to ImageNet-initialized ViT-Large baseline, Arctic MAE pretraining produced consistent improvements in foreground mean F1 scores of 0. 87, 0. 72, 0. 93, and 0.
87, for infrastructure, IWP, RTS, and TCNs, with approximately 5-8 percentage increase. The proposed model also outperformed Prithvi-EO-2. 0 in all downstream comparisons, with the smallest gain corresponding to at least a 15 percentage improvement mean F1, suggesting that domain-specific self-supervised pretraining on curated Arctic VHSR imagery provides more transferable representations for fine-scale Arctic mapping than a general-purpose Earth observation foundation model.
These results demonstrate that optimizing the pretraining data distribution at regional scale, while keeping the architecture and MAE objective fixed, can produce a reusable Arctic-domain encoder for multiple VHSR remote sensing applications.
Reader Mode unavailable (could not extract clean content).
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →Evi-Steer: Learning to Steer Biomedical Vision-Language Models through Efficient and Generalizable Evidential Tuning
Evi-Steer introduces a novel evidential tuning framework for BiomedCLIP, enabling efficient fine-tuning with only 0.11% parameter updates. It significantly enhances performance in few-shot learning and domain shifts across 15 biomedical imaging datasets, demonstrating robustness for clinical applications.