Unsupervised Learning of Visual Features by Contrasting Cluster Assignments

Open paper
Benchmark Model Rank Results
image-classification-on-inaturalist-2018ResNet-50#52Top-1 Accuracy: 48.6
image-classification-on-omnibenchmarkSwAV#9Average Top-1 Accuracy: 38.3
image-classification-on-places205SwAV#10Top 1 Accuracy: 56.7%
image-classification-on-places205ResNet-50 (Supervised)#14Top 1 Accuracy: 53.2%
self-supervised-image-classification-on-imagenetSwAV (ResNet-50 x5)#41Top 1 Accuracy: 78.5%Number of Params: 586M
self-supervised-image-classification-on-imagenetSwAV (ResNet-50 x2)#50Top 1 Accuracy: 77.3%Number of Params: 94M
self-supervised-image-classification-on-imagenetSwAV (ResNet-50)#67Top 1 Accuracy: 75.3%Number of Params: 24M
self-supervised-image-classification-on-imagenetDeepCluster-v2 (ResNet-50)#70Top 1 Accuracy: 75.2%Number of Params: 24M
self-supervised-image-classification-on-imagenet-finetunedSwAV (ResNeXt-101-32x16d)#54Top 1 Accuracy: 82.0%Number of Params: 193M
self-supervised-image-classification-on-imagenet-finetunedSwAV (Resnet-50)#60Top 1 Accuracy: 77.8%Number of Params: 182M
semi-supervised-image-classification-on-imagenet-1-labeled-dataSwAV (ResNet-50)#43Top 1 Accuracy: 53.9%Top 5 Accuracy: 78.5

Rank counts only results with a code link.