A survey of image labelling for computer vision applications

Supervised machine learning methods for image analysis require large amounts of labelled training data to solve computer vision problems. The recent rise deep algorithms recognising content has led the emergence many ad-hoc labelling tools. With this survey, we capture and systematise commonalities as well distinctions between existing software. We perform a structured literature review compile underlying concepts features software such annotation expressiveness degree automation. structure manual task by its organisation work, user interface design options, support techniques derive systematisation schema survey. Applying it available body literature, enabled us uncover several application archetypes key domains retrieval or instance identification in healthcare television.