List your company
ResearchFormerly WizWand

Medical Image Segmentation on SEGRAP

73.1DSCAdaptformer + SAM-COBOT
7.99624.89841.858.702May 17, 2023Jun 18, 2023Jul 21, 2023Aug 22, 2023Sep 24, 2023Oct 26, 2023Nov 28, 2023
Updated 8mo ago

Evaluation results

MethodLinks
Adaptformer + SAM-COBOTTrainable Parameters (K)=324.0, Backbone=ViT-Base2023.1173.1
AdaptformerTrainable Parameters (K)=322.7, Backbone=ViT-Base2023.1172.1
LoRA + SAM-COBOTTrainable Parameters (K)=148.3, Backbone=ViT-Base2023.1170.1
LoRATrainable Parameters (K)=147.4, Backbone=ViT-Base2023.1168.7
LightweightTrainable Parameters (K)=0, Backbone=ViT-Base2023.1167.8
One-Promptsetting=segment everything, mode=zero-shot2023.0562.2
MedSAMsetting=segment everything, mode=zero-shot2023.0552.3
MSAsetting=segment everything, mode=zero-shot2023.0547.3
SAM-Med2Dsetting=segment everything, mode=zero-shot2023.0543.5
MedSegDiffsetting=segment everything, mode=zero-shot2023.0534.7
nnUNetsetting=segment everything, mode=zero-shot2023.0526.8
TransUNetsetting=segment everything, mode=zero-shot2023.0525.5
Swin-UNetrsetting=segment everything, mode=zero-shot2023.0523.6
FreezeTrainable Parameters (K)=0, Backbone=ViT-Base2023.1110.5