Unsupervised learning of geometrical features from images by explicit group actions enforcement

In this work we propose an autoencoder architecture capable of automatically learning meaningful geometric features of objects in images, achieving a disentangled representation of 2D objects. It is made of a standard dense autoencoder that captures the deep features identifying the shapes and an additional encoder that extracts geometric latent variables regressed in an unsupervised manner. These are then used to apply a transformation on the output of the \textit{deep features} decoder. The promising results show that this approach performs better than a non-constrained model having more degrees of freedom.