ZegCLIP: Towards Adapting CLIP for Zero-shot Semantic Segmentation

Benchmark Model Rank Results
zero-shot-semantic-segmentation-on-coco-stuffZegCLIP#4Transductive Setting hIoU: 48.5Inductive Setting hIoU: 40.8
zero-shot-semantic-segmentation-on-pascal-vocZegCLIP#4Transductive Setting hIoU: 91.1Inductive Setting hIoU: 84.3