In Defense of Lazy Visual Grounding for Open-Vocabulary Semantic Segmentation

Benchmark Model Rank Results
open-vocabulary-semantic-segmentation-on-1LaVG#20mIoU: 34.7
open-vocabulary-semantic-segmentation-on-2LaVG#19mIoU: 15.8
open-vocabulary-semantic-segmentation-on-5LaVG#15mIoU: 82.5
open-vocabulary-semantic-segmentation-on-cocoLaVG#5mIoU: 23.2