The online fashion industry is growing fast and with it, the need for advanced systems able to automatically solve different tasks in an accurate way. With the rapid advance of digital technologies, Deep Learning has played an important role in Computational Aesthetics, an interdisciplinary area that tries to bridge fine art, design, and computer science. Specifically, Computational Aesthetics aims to automatize human aesthetic judgments with computational methods. In this thesis, we focus on three applications of computer vision in fashion, and we discuss how Computational Aesthetics helps solve them accurately.