Several research areas in the humanities and social sciences could potentially benefit from state-of-the-art machine learning technologies. In the field of computer vision, there are exemplary face detection (FD) and face recognition (FR) algorithms that could provide support on several levels. We discuss two application scenarios where the deployment of trained networks can be used to generate further information. We show how FD can be used to recognize scene types such as dialogue and speech on non-photographic images such as Emblematica. The second part shows another application scenario where FR can be used to combine or link images in research datasets with authority records by finding personalities from the Wikidata dataset.