The field of computer vision is rapidly evolving, particularly in the area of unsupervised deep learning. Over the past year or so there have been many new and exciting methods developed to both represent and generate images in an automated fashion, but the field is evolving so rapidly that it can be hard to keep track of all these methods. I recently gave a research talk to the Styling Algorithms team here at Stitch Fix on the current state of the art (as I see it) in unsupervised computer vision research. It was by no means comprehensive, but more of a survey of of interesting methods I thought might be applicable to a problem I have been working on recently: how does one disentangle attributes at the level of a latent image representation?
I particularly find the semi-supervised results in the Adversarial Auto-encoder models to be very compelling. The slide deck below covers a few different models that have come out relatively recently. It starts by reviewing Variational Auto-encoders, a more thorough treatment of which you can find in the original paper by Kingma and Welling or my previous post in which I apply their model to our clothing inventory. Then I review Generative Adversarial Networks (GAN), the hybrid algorithm VAE/GAN, Generative Moment-Matching Networks (GMMN), and Adversarial Auto-encoders. The slides below briefly go into the workings of each and some interesting results that fall out of them.
