Multimodal Classification on FOOD-101 Hard (test)
93.4AccuracyMMBT-Large
Evaluation Results
| Method | Links | |
|---|---|---|
| MMBT-LargeModel Scale=Large2019.09 | 93.4 | |
| ViLBertFine-tuning Source=Standard pre-trained2019.09 | 92.9 | |
| MMBTModel Scale=Base2019.09 | 92.4 | |
| ViLBert-VQAFine-tuning Source=VQA2019.09 | 92.4 | |
| ViLBert-Flickr30kFine-tuning Source=Flickr30k2019.09 | 92.2 | |
| ViLBert-VCRFine-tuning Source=VCR2019.09 | 92.1 | |
| ViLBert-RefcocoFine-tuning Source=Refcoco2019.09 | 92.1 |