Self-supervised Image-specific Prototype Exploration for Weakly Supervised Semantic Segmentation

Open paper
Benchmark Model Rank Results
weakly-supervised-semantic-segmentation-onSIPE (DeepLabV2-ResNet101, no saliency)#52Mean IoU: 68.8
weakly-supervised-semantic-segmentation-on-1SIPE (DeepLabV2-ResNet101, no saliency)#46Mean IoU: 69.7
weakly-supervised-semantic-segmentation-on-4SIPE(ResNet-38, no saliency, no RW)#21mIoU: 43.6

Rank counts only results with a code link.