Preprint
Aug 2026
Semantic-Spatial Discriminability Enhancement for Generalized Visual Grounding
A novel Semantic-Spatial Discriminability Enhancement (SSDE) framework for generalized visual grounding is proposed, which aims to enhance the discriminative ability on fine-grained semantics and spatial localization, improving both cross-modal understanding and instance-level grounding.
Kai Lei, Xu-Yao Zhang
· 0 citations