Image Segmentation on SONAR (mIoU)
84.9mIoUAdaptformer + SAM-COBOT
Evaluation Results
| Method | Links | |
|---|---|---|
| Adaptformer + SAM-COBOTTrainable Parameters (K)=324.0, Backbone=ViT-Base2023.11 | 84.9 | |
| LoRA + SAM-COBOTTrainable Parameters (K)=148.3, Backbone=ViT-Base2023.11 | 84.6 | |
| AdaptformerTrainable Parameters (K)=322.7, Backbone=ViT-Base2023.11 | 84.1 | |
| LoRATrainable Parameters (K)=147.4, Backbone=ViT-Base2023.11 | 82.8 | |
| LightweightTrainable Parameters (K)=0, Backbone=ViT-Base2023.11 | 79.8 | |
| FreezeTrainable Parameters (K)=0, Backbone=ViT-Base2023.11 | 34.5 |