Feature Guided Masked Autoencoder for Self-supervised Learning in Remote Sensing

Open paper
Benchmark Model Rank Results
image-classification-on-eurosat-sarFG-MAE (ViT-S/16)#1Overall Accuracy: 85.9
image-classification-on-eurosat-sarMAE (ViT-S/16)#2Overall Accuracy: 81.0
image-classification-on-eurosat-sarViT-S/16#3Overall Accuracy: 78.4
multi-label-image-classification-on-2FG-MAE (ViT-S/16)#2mAP (micro): 89.3F1 Score: 80.8
multi-label-image-classification-on-2MAE (ViT-S/16)#4mAP (micro): 88.6F1 Score: 79.9
multi-label-image-classification-on-2ViT-S/16#5mAP (micro): 87.8F1 Score: 78.9
multi-label-image-classification-on-3FG-MAE (ViT-S/16)#1mAP (micro): 82.7
multi-label-image-classification-on-3MAE (ViT-S/16)#2mAP (micro): 81.3
multi-label-image-classification-on-3ViT-S/16#3mAP (micro): 79.5

Rank counts only results with a code link.