CLIP is Also an Efficient Segmenter: A Text-Driven Approach for Weakly Supervised Semantic Segmentation

Benchmark Model Rank Results
weakly-supervised-semantic-segmentation-onCLIP-ES(DeepLabV2-ResNet101)#14Mean IoU: 73.8
weakly-supervised-semantic-segmentation-on-1CLIP-ES(DeepLabV2-ResNet101)#13Mean IoU: 73.9
weakly-supervised-semantic-segmentation-on-4CLIP-ES(DeepLabV2-ResNet101)#13mIoU: 45.4