Dense Cross-Query-and-Support Attention Weighted Mask Aggregation for Few-Shot Segmentation

Open paper
Benchmark Model Rank Results
few-shot-semantic-segmentation-on-coco-20i-1DCAMA (Swin-B)#7Mean IoU: 50.9FB-IoU: 73.2
few-shot-semantic-segmentation-on-coco-20i-1DCAMA (ResNet-101)#37Mean IoU: 43.5FB-IoU: 69.9learnable parameters (million): 47.7
few-shot-semantic-segmentation-on-coco-20i-1DCAMA (ResNet-50)#38Mean IoU: 43.3FB-IoU: 69.5
few-shot-semantic-segmentation-on-coco-20i-2-1DCAMA (Swin-B)#3mIoU: 31.7
few-shot-semantic-segmentation-on-coco-20i-5DCAMA (Swin-B)#7Mean IoU: 58.3FB-IoU: 76.9
few-shot-semantic-segmentation-on-coco-20i-5DCAMA (ResNet-101)#25Mean IoU: 51.9FB-IoU: 73.3learnable parameters (million): 47.7
few-shot-semantic-segmentation-on-coco-20i-5DCAMA (ResNet-50)#43Mean IoU: 48.3FB-IoU: 71.7learnable parameters (million): 47.7
few-shot-semantic-segmentation-on-fss-1000-1DCAMA (Swin-B)#6Mean IoU: 90.1FB-IoU: 93.8
few-shot-semantic-segmentation-on-fss-1000-1DCAMA (ResNet-101)#11Mean IoU: 88.3FB-IoU: 92.4
few-shot-semantic-segmentation-on-fss-1000-1DCAMA (ResNet-50)#12Mean IoU: 88.2FB-IoU: 92.5
few-shot-semantic-segmentation-on-fss-1000-5DCAMA (Swin-B)#7Mean IoU: 90.4FB-IoU: 94.1
few-shot-semantic-segmentation-on-fss-1000-5DCAMA (ResNet-101)#10Mean IoU: 89.1FB-IoU: 93.1
few-shot-semantic-segmentation-on-fss-1000-5DCAMA (ResNet-50)#12Mean IoU: 88.8FB-IoU: 92.9
few-shot-semantic-segmentation-on-pascal-5i-1DCAMA (Swin-B)#8Mean IoU: 69.3FB-IoU: 78.5
few-shot-semantic-segmentation-on-pascal-5i-1DCAMA (ResNet-50)#47Mean IoU: 64.6FB-IoU: 75.7
few-shot-semantic-segmentation-on-pascal-5i-1DCAMA (ResNet-101)#89FB-IoU: 77.6
few-shot-semantic-segmentation-on-pascal-5i-5DCAMA (Swin-B)#5Mean IoU: 74.9FB-IoU: 82.9
few-shot-semantic-segmentation-on-pascal-5i-5DCAMA (ResNet-50)#48Mean IoU: 68.5FB-IoU: 79.5
few-shot-semantic-segmentation-on-pascal-5i-5DCAMA (ResNet-101)#49Mean IoU: 68.3FB-IoU: 80.8

Rank counts only results with a code link.