IMAGINATOR: Pre-Trained Image+Text Joint Embeddings using Word-Level Grounding of Images

Open in new window