Large Scale Image Annotation: Learning to Rank with Joint Word-Image Embeddings

   Abstract