Visual and Textual Prior Guided Mask Assemble for Few-Shot Segmentation and Beyond

Benchmark Model Rank Results
few-shot-semantic-segmentation-on-coco-20i-1-shotPGMA-Net (ResNet-101)–Mean IoU: 59.4FB-IoU: 78.5
few-shot-semantic-segmentation-on-coco-20i-1-shotPGMA-Net (ResNet-50)–Mean IoU: 54.3FB-IoU: 75.8
few-shot-semantic-segmentation-on-coco-20i-5-shotPGMA-Net (ResNet-101)–Mean IoU: 61.8FB-IoU: 79.4
few-shot-semantic-segmentation-on-coco-20i-5-shotPGMA-Net (ResNet-50)–Mean IoU: 57.1FB-IoU: 76.7
few-shot-semantic-segmentation-on-pascal-5i-1-shotPGMA-Net (ResNet-101)–Mean IoU: 77.6FB-IoU: 86.2
few-shot-semantic-segmentation-on-pascal-5i-1-shotPGMA-Net (ResNet-50)–Mean IoU: 74.1FB-IoU: 83.5
few-shot-semantic-segmentation-on-pascal-5i-1-shotPGMA-Net (ViT-B/16)–Mean IoU: 74.1FB-IoU: 82.1
few-shot-semantic-segmentation-on-pascal-5i-5-shotPGMA-Net (ResNet-101)–Mean IoU: 78.6FB-IoU: 86.9
few-shot-semantic-segmentation-on-pascal-5i-5-shotPGMA-Net (ResNet-50)–Mean IoU: 75.2FB-IoU: 84.2
few-shot-semantic-segmentation-on-pascal-5i-5-shotPGMA-Net (ViT-B/16)–Mean IoU: 74.6FB-IoU: 82.5