Preprint
Jul 2026
Beyond Isolated Objects: Relationship-aware Open Vocabulary Scene Understanding via 3D Scene Graph Analysis
An Adaptive Gated Dual-Stream Contextual GAT that separates dense geometric features and semantic CLIP embeddings, performs edge-guided message passing, and adaptively fuses complementary semantics is introduced that separates dense geometric features and semantic CLIP embeddings, and adaptively fuses complementary semantics.
Xianhao Chen, Jiarui Hu, Yuanbo Yang et al.
· 0 citations