AI RESEARCH
CAGE-SGG: Counterfactual Active Graph Evidence for Open-Vocabulary Scene Graph Generation
arXiv CS.CV
•
ArXi:2604.22274v1 Announce Type: new Open-vocabulary scene graph generation (SGG) aims to describe visual scenes with flexible and fine-grained relation phrases beyond a fixed predicate vocabulary. While recent vision-language models greatly expand the semantic coverage of SGG, they also