Reducing Information Bottleneck for Weakly Supervised Semantic Segmentation

Benchmark Model Rank Results
weakly-supervised-semantic-segmentation-onRIB+Sal (DeepLabV2-ResNet101)#42Mean IoU: 70.2
weakly-supervised-semantic-segmentation-onRIB (DeepLabV2-ResNet101)#57Mean IoU: 68.3
weakly-supervised-semantic-segmentation-on-1RIB+Sal (DeepLabV2-ResNet101)#44Mean IoU: 70.0
weakly-supervised-semantic-segmentation-on-1RIB (DeepLabV2-ResNet101)#52Mean IoU: 68.6
weakly-supervised-semantic-segmentation-on-4RIB (DeepLabV2-ResNet101, No Saliency)#20mIoU: 43.8