Advancing Visual Grounding with Scene Knowledge: Benchmark and Method

Open in new window