SatDINO, a DINO-based self-supervised ViT pretrained on fMoW-RGB, produces stronger frozen-feature representations (kNN, linear probing) than MAE-based models like Scale-MAE and SatMAE, with competitive fine-tuning and segmentation results.
Geo- clip: Clip-inspired alignment between locations and images for effective worldwide geo-localization,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
SatDINO: A Deep Dive into Self-Supervised Pretraining for Remote Sensing
SatDINO, a DINO-based self-supervised ViT pretrained on fMoW-RGB, produces stronger frozen-feature representations (kNN, linear probing) than MAE-based models like Scale-MAE and SatMAE, with competitive fine-tuning and segmentation results.