Image-to-Text Retrieval on FashionGen (test)
73.14R@1FashionSAP
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| FashionSAP2023.04 | 73.14 | 92.8 | 96.87 | 71.63 | |
| FashionSAPBackbone=ViT-B162023.04 | 71.14 | 92.21 | 96.52 | 70.11 | |
| FashionSAPBackbone=ResNet-502023.04 | 67.23 | 91.3 | 96.41 | 65.67 | |
| FAME-ViLBackbone=ViT-B/16, Learning Strategy=MTL, Test Protocol=random 1002023.03 | 65.94 | 91.92 | 97.22 | 83.14 | |
| FashionViLBackbone=ResNet-50, Learning Strategy=STL, Test Protocol=random 1002023.03 | 65.54 | 91.34 | 96.3 | 82.6 | |
| FashionViL2023.04 | 65.54 | 91.34 | 96.3 | 63.71 | |
| FAME-ViL(ST)Backbone=ViT-B/16, Learning Strategy=STL, Test Protocol=random 1002023.03 | 65.02 | 90.96 | 96.2 | 82.61 | |
| ALBEF2023.04 | 63.97 | 88.92 | 94.41 | 62.25 | |
| FashionViL(vit)Backbone=ViT-B/16, Learning Strategy=STL, Test Protocol=random 1002023.03 | 63.74 | 90.02 | 95.98 | 81.61 | |
| CommerceMM2023.04 | 41.6 | 64 | 72.8 | 40.6 | |
| EI-CLIPLearning Strategy=STL, Test Protocol=random 1002023.03 | 38.7 | 72.2 | 84.25 | 65.02 | |
| EI-CLIP2023.04 | 38.7 | 72.2 | 84.25 | 39.38 | |
| MVLTLearning Strategy=STL, Test Protocol=random 1002023.03 | 33.1 | 77.2 | 91.1 | 67.25 | |
| KaleidoBERTLearning Strategy=STL, Test Protocol=random 1002023.03 | 27.99 | 60.09 | 68.37 | 53.25 | |
| KaleidoBERT2023.04 | 27.99 | 60.09 | 68.37 | 30.94 | |
| FashionBERTLearning Strategy=STL, Test Protocol=random 1002023.03 | 23.96 | 46.31 | 52.12 | 41.89 | |
| FashionBERT2023.04 | 23.96 | 46.31 | 52.12 | 25.36 | |
| OSCARLearning Strategy=STL, Test Protocol=random 1002023.03 | 23.39 | 44.67 | 52.55 | 41.92 | |
| OSCAR2023.04 | 23.39 | 44.67 | 52.55 | 24.25 | |
| Image-BERT2023.04 | 22.76 | 41.89 | 50.77 | 23.77 | |
| VILBERT2023.04 | 20.97 | 40.49 | 48.21 | 21.05 | |
| VL-BERT2023.04 | 19.26 | 39.9 | 46.05 | 20.95 |