A 15-million-image sub-meter remote sensing dataset built from Jilin-1 imagery and a multi-scale self-supervised ViT framework achieve benchmark results comparable to or better than prior remote sensing foundation models.
Full Convolution Neural Network Combined with Contextual Feature Representation for Cropland Extraction from High-Resolution Remote Sensing Images,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
CGEarthEye:A High-Resolution Remote Sensing Vision Foundation Model Based on the Jilin-1 Satellite Constellation
A 15-million-image sub-meter remote sensing dataset built from Jilin-1 imagery and a multi-scale self-supervised ViT framework achieve benchmark results comparable to or better than prior remote sensing foundation models.