Learning to Generate Text-Grounded Mask for Open-World Semantic Segmentation from Only Image-Text Pairs.

Junbum Cha, Jonghwan Mun, Byungseok Roh

Browse the full CVPR paper archive.