Unsupervised Image Classification on ImageNet Superclasses (10 classes) WordNet hierarchy (val)
85.7AccuracySelf-Classifier
Evaluation results
| Method | Links | |
|---|---|---|
| Self-ClassifierBackbone=ResNet-502021.03 | 85.7 | |
| SCANBackbone=ResNet-50, Pre-training=ImageNet-pretrained2021.03 | 85.3 | |
| OBoWBackbone=ResNet-50, Evaluation Protocol=k-means classifier2021.03 | 83.9 | |
| BarlowTBackbone=ResNet-50, Evaluation Protocol=k-means classifier2021.03 | 80.2 | |
| MoCoV2Backbone=ResNet-50, Evaluation Protocol=k-means classifier2021.03 | 80 | |
| DINOBackbone=ResNet-50, Evaluation Protocol=k-means classifier2021.03 | 79.7 | |
| SwAVBackbone=ResNet-50, Evaluation Protocol=k-means classifier2021.03 | 79.1 | |
| SeLaBackbone=ResNet-50, Pre-training=ImageNet-pretrained2021.03 | 55.2 |