MSDNet: Multi-Scale Decoder for Few-Shot Semantic Segmentation via Transformer-Guided Prototyping

Benchmark Model Rank Results
few-shot-semantic-segmentation-on-coco-20i-pascal-voc-1-shotMSDNet (ResNet-101)#1Mean IoU: 73.9
few-shot-semantic-segmentation-on-coco-20i-pascal-voc-1-shotMSDNet (ResNet-50)#2Mean IoU: 72.1
few-shot-semantic-segmentation-on-coco-20i-pascal-voc-5-shotMSDNet (ResNet-101)#5Mean IoU: 76.4
few-shot-semantic-segmentation-on-coco-20i-pascal-voc-5-shotMSDNet (ResNet-50)#6Mean IoU: 74.2
few-shot-semantic-segmentation-on-coco-20i-1-shotMSDNet (ResNet-101)#12Mean IoU: 48.5FB-IoU: 71.3learnable parameters (million): 1.5
few-shot-semantic-segmentation-on-coco-20i-1-shotMSDNet (ResNet-50)#22Mean IoU: 46.5FB-IoU: 70.4learnable parameters (million): 1.5
few-shot-semantic-segmentation-on-coco-20i-5-shotMSDNet (ResNet-101)#14Mean IoU: 55.3FB-IoU: 75.1learnable parameters (million): 1.5
few-shot-semantic-segmentation-on-coco-20i-5-shotMSDNet (ResNet-50)#16Mean IoU: 54.5FB-IoU: 74.5learnable parameters (million): 1.5
few-shot-semantic-segmentation-on-pascal-5i-5-shotMSDNet (ResNet-101)#38Mean IoU: 70.8FB-IoU: 85.0learnable parameters (million): 1.5
few-shot-semantic-segmentation-on-pascal-5i-5-shotMSDNet (ResNet-50)#54Mean IoU: 68.7FB-IoU: 82.1learnable parameters (million): 1.5
few-shot-semantic-segmentation-on-pascal-5i-1-shotMSDNet (ResNet-101)#54Mean IoU: 64.7FB-IoU: 77.3learnable parameters (million): 1.5
few-shot-semantic-segmentation-on-pascal-5i-1-shotMSDNet (ResNet-50)#61Mean IoU: 64.3FB-IoU: 77.1learnable parameters (million): 1.5