Cost Aggregation with 4D Convolutional Swin Transformer for Few-Shot Segmentation

Open paper
Benchmark Model Rank Results
few-shot-semantic-segmentation-on-coco-20i-1VAT (ResNet-101)#52Mean IoU: 41.3FB-IoU: 68.8
few-shot-semantic-segmentation-on-coco-20i-5VAT (ResNet-101)#46Mean IoU: 47.9FB-IoU: 72.4
few-shot-semantic-segmentation-on-fss-1000-1VAT (ResNet-101)#4Mean IoU: 90.3FB-IoU: 94
few-shot-semantic-segmentation-on-fss-1000-1VAT (ResNet-50)#7Mean IoU: 90.1FB-IoU: 93.8
few-shot-semantic-segmentation-on-fss-1000-5VAT (ResNet-101)#3Mean IoU: 90.8FB-IoU: 94.4
few-shot-semantic-segmentation-on-fss-1000-5VAT (ResNet-50)#4Mean IoU: 90.7FB-IoU: 94.2
few-shot-semantic-segmentation-on-pascal-5i-1VAT (ResNet-101)#18Mean IoU: 67.9FB-IoU: 79.6
few-shot-semantic-segmentation-on-pascal-5i-1VAT (ResNet-50)#40Mean IoU: 65.5FB-IoU: 77.8
few-shot-semantic-segmentation-on-pascal-5i-5VAT (ResNet-101)#16Mean IoU: 72FB-IoU: 83.2
few-shot-semantic-segmentation-on-pascal-5i-5VAT (ResNet-50)#36Mean IoU: 70.1FB-IoU: 80.9
semantic-correspondence-on-pf-pascalVAT (ECCV)#8PCK: 92.3
semantic-correspondence-on-pf-willowVAT (ECCV)#2PCK: 81.6
semantic-correspondence-on-spair-71kVAT (ECCV)#11PCK: 55.5

Rank counts only results with a code link.