MoCo-v2 (ResNet18, fine tune)

Reported on 3 benchmarks across 2 tasks · 1 paper · 2 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision3 results

Multi-Label Image ClassificationonBigEarthNet
mAP (micro)· 2022-06-27
89.3
best: 91.8 (MoCo-v2 (ResNet50, fine tune))
SOTA
Self-supervised Learning in Remote Sensing: A Review arXiv:2206.13188
Image ClassificationonBigEarthNet
mAP (micro)· 2022-06-27
89.3
best: 91.8 (MoCo-v2 (ResNet50, fine tune))
SOTA
Self-supervised Learning in Remote Sensing: A Review arXiv:2206.13188
Image ClassificationonEuroSAT
Accuracy (%)· uses extra data· 2022-06-27
98.9
best: 99.41 (DeepEnsembling)
Self-supervised Learning in Remote Sensing: A Review arXiv:2206.13188