Masked Autoencoding (MAE) Outperforms 1 Joint-Embedding Prediction (JEPA) for Frozen-Probe 2 Very-High-Resolution Landslide Segmentation
· 0 citations
· 50 references