Fetching the paper…

Towards Unseen Triples: Effective Text-Image-joint Learning for Scene Graph Generation · Around