Image Classification on ImageNet-Sketch (IN-S) (test)
37.1Top-1 AccuracyMultiMAE
Evaluation results
| Method | Links | |
|---|---|---|
| MultiMAEBackbone=ViT-B, Modality=RGB-only, Training=Fine-tuned2022.04 | 37.1 | |
| MoCo-v3Backbone=ViT-B, Modality=RGB-only, Training=Fine-tuned2022.04 | 35.6 | |
| MAEBackbone=ViT-B, Modality=RGB-only, Training=Fine-tuned2022.04 | 35.5 | |
| DINOBackbone=ViT-B, Modality=RGB-only, Training=Fine-tuned2022.04 | 35.4 | |
| SupervisedBackbone=ViT-B, Modality=RGB-only, Training=Fine-tuned2022.04 | 31.4 |