... that generate image captions for news articles A key aspect of our approach is to allow both the visual and textual modalities to influence the generation task This is achieved through an image ... Formulation We formulate image caption generation as follows Given an image I, and a related knowledge database κ, create a natural language description C which captures the main content of the image ... Instead of relying on manual annotation or background ontological information we exploit a multimodal database of news articles, images, and their captions The latter is admittedly noisy, yet can...