Image Classification on Dped (test)
98.7Top-1 AccVGG-B(P)
Evaluation results
| Method | Links | |
|---|---|---|
| VGG-B(P)Backbone=VGG-B, Pre-training=pre-trained on ImageNet, #par=82017.05 | 98.7 | |
| DAN_imagenetBackbone=VGG-B, Controller Initialization=ImageNet, #par=2.762017.05 | 98.7 | |
| DAN_imagenet+sketchBackbone=VGG-B, Controller Initialization=ImageNet + Sketch, #par=3.322017.05 | 98.7 | |
| DAN_sketchBackbone=VGG-B, Controller Initialization=sketch, #par=2.542017.05 | 94 | |
| DAN_caltech-256Backbone=VGG-B, Controller Initialization=caltech-256, #par=2.542017.05 | 93.8 | |
| VGG-B(S)Backbone=VGG-B, Pre-training=trained from scratch, #par=82017.05 | 92.9 | |
| DAN_noiseBackbone=VGG-B, Controller Initialization=random weights, #par=1.762017.05 | 91.3 |