Learning Disentangled Representations in the Imaging Domain

Disentangled representation learning has been proposed as an approach to learning general representations even in the absence of, or with limited, supervision. A good general representation can be fine-tuned for new target tasks using modest amounts of data, or used directly in unseen domains achieving remarkable performance in the corresponding task. This alleviation of the data and annotation requirements offers tantalising prospects for applications in computer vision and healthcare. In this tutorial paper, we motivate the need for disentangled representations, present key theory, and detail practical building blocks and criteria for learning such representations. We discuss applications in medical imaging and computer vision emphasising choices made in exemplar key works. We conclude by presenting remaining challenges and opportunities.