Кластеризация изображения по визуальному подобию с помощью автоэнкодера

Abstract

The work considers the approach to solving the problem of finding similar images by visual similarity using neural networks on previously unmarked data. The solution comes down to building special architecture of the neural network - autoencoder, through which extracted high-level features from images. Search for the nearest items implemented by entered metric in the formed feature space. This metric can be applied to the classification task using pre-clustering.