Transferable Decoding with Visual Entities for Zero-Shot Image Captioning

Open in new window