Multimodal Deep Learning in Semantic Image Segmentation: A Review

In recent years, there has been a lot of research in the area of semantic image segmentation, which involves breaking down an image into its discrete components, such that humans can give meaning to its contents. From the humble beginnings of image search using human-provided captions, content-based image retrieval has come a long way. Yet, areas of research and improvement are far from diminishing. In this paper we will take a look at how multi-modal approaches to semantic image segmentation are setting the new standard in image search and retrieval.