Uncertainty Estimation via Response Scaling for Pseudo-mask Noise Mitigation in Weakly-supervised Semantic Segmentation

Open paper
Benchmark Model Rank Results
weakly-supervised-semantic-segmentation-onURN(Res2Net-101, no saliency, no RW)#26Mean IoU: 71.2
weakly-supervised-semantic-segmentation-onURN(ScaleNet-101, no saliency, no RW)#43Mean IoU: 70.1
weakly-supervised-semantic-segmentation-onURN(ResNet-101, no saliency, no RW)#46Mean IoU: 69.5
weakly-supervised-semantic-segmentation-onURN(ResNet-38, no saliency, no RW)#50Mean IoU: 69.4
weakly-supervised-semantic-segmentation-on-1URN(Res2Net-101, no saliency, no RW)#27Mean IoU: 71.5
weakly-supervised-semantic-segmentation-on-1URN(ScaleNet-101, no saliency, no RW)#33Mean IoU: 70.8
weakly-supervised-semantic-segmentation-on-1URN(ResNet-38, no saliency, no RW)#38Mean IoU: 70.6
weakly-supervised-semantic-segmentation-on-1URN(ResNet-101, no saliency, no RW)#45Mean IoU: 69.7
weakly-supervised-semantic-segmentation-on-4URN(Res2Net-101, no saliency, no RW)#24mIoU: 41.5
weakly-supervised-semantic-segmentation-on-4URN(ScaleNet-101, no saliency, no RW)#25mIoU: 40.8
weakly-supervised-semantic-segmentation-on-4URN(ResNet-101, no saliency, no RW)#26mIoU: 40.7
weakly-supervised-semantic-segmentation-on-4URN(ResNet-38, no saliency, no RW)#27mIoU: 40.5

Rank counts only results with a code link.