Read Before Grounding: Scene Knowledge Visual Grounding via Multi-step Parsing.

Haixiang Zhu, Lixian Su, ShuangMing Mao, Jing Ye

Browse the full COLING paper archive.