Semantic Segmentation on FoodSeg103 1.0 (test)
43.9mIoUReLeM-SeTR
Evaluation results
| Method | Links | ||
|---|---|---|---|
| ReLeM-SeTRReLeM Variant=LSTM, Backbone=ViT-16/B, Model Size=723M2021.05 | 43.9 | 57 | |
| ReLeM-SeTRReLeM Variant=Transformer, Backbone=ViT-16/B, Model Size=723M2021.05 | 43.2 | 55.7 | |
| SeTRBackbone=ViT-16/B, Model Size=723M2021.05 | 41.3 | 52.7 | |
| ReLeM-CCNetReLeM Variant=LSTM, Backbone=ResNet-50, Model Size=381M2021.05 | 36.8 | 47.4 | |
| ReLeM-CCNetReLeM Variant=Transformer, Backbone=ResNet-50, Model Size=381M2021.05 | 36 | 46.5 | |
| CCNetBackbone=ResNet-50, Model Size=381M2021.05 | 35.5 | 45.3 | |
| ReLeM-FPNReLeM Variant=LSTM, Backbone=ResNet-50, Model Size=218M2021.05 | 29.1 | 39.8 | |
| ReLeM-FPNReLeM Variant=Transformer, Backbone=ResNet-50, Model Size=218M2021.05 | 28.9 | 39.7 | |
| FPNBackbone=ResNet-50, Model Size=218M2021.05 | 27.8 | 38.2 |