Image Classification on Sketch (test)
96.15Top-1 AccSL-MLP
Evaluation results
| Method | Links | |
|---|---|---|
| SL-MLPEvaluation Protocol=Full-data finetuning, Pre-trained Dataset=ImageNet-1K, Pre-training Epochs=3002021.12 | 96.15 | |
| SLEvaluation Protocol=Full-data finetuning, Pre-trained Dataset=ImageNet-1K, Pre-training Epochs=3002021.12 | 95.9 | |
| SupConEvaluation Protocol=Full-data finetuning, Pre-trained Dataset=ImageNet-1K, Pre-training Epochs=3002021.12 | 95.72 | |
| SupCon w/o MLPEvaluation Protocol=Full-data finetuning, Pre-trained Dataset=ImageNet-1K, Pre-training Epochs=3002021.12 | 95.68 | |
| NETTAILORBackbone=ResNet-50, Input Resolution=224x224, Number of Parameters=15.1M, FLOPs=3.44G2019.06 | 80.48 | |
| PiggybackBackbone=ResNet-50, Input Resolution=224x224, Number of Parameters=24.3M, FLOPs=4.11G2019.06 | 79.91 | |
| PackNet (Reversed)Backbone=ResNet-50, Input Resolution=224x224, Order=Sketch, WikiArt, Flowers, Cars, CUB, Number of Parameters=23.9M, FLOPs=4.11G2019.06 | 78.7 | |
| PackNet (Forward)Backbone=ResNet-50, Input Resolution=224x224, Order=CUB, Cars, Flowers, WikiArt, Sketch, Number of Parameters=23.9M, FLOPs=4.11G2019.06 | 76.17 | |
| VGG-B(S)Backbone=VGG-B, Pre-training=trained from scratch, #par=82017.05 | 69.2 | |
| DAN_sketchBackbone=VGG-B, Controller Initialization=sketch, #par=2.542017.05 | 69.2 | |
| DAN_imagenet+sketchBackbone=VGG-B, Controller Initialization=ImageNet + Sketch, #par=3.322017.05 | 69.2 | |
| VGG-B(P)Backbone=VGG-B, Pre-training=pre-trained on ImageNet, #par=82017.05 | 65.4 | |
| DAN_imagenetBackbone=VGG-B, Controller Initialization=ImageNet, #par=2.762017.05 | 63.2 | |
| Feature ExtractionBackbone=ResNet-50, Input Resolution=224x224, Number of Parameters=23.9M, FLOPs=4.11G2019.06 | 50.86 | |
| DAN_caltech-256Backbone=VGG-B, Controller Initialization=caltech-256, #par=2.542017.05 | 49.4 | |
| DAN_noiseBackbone=VGG-B, Controller Initialization=random weights, #par=1.762017.05 | 42.7 |