Decoupling Global and Local Representations via Invertible Generative Flows

A new generative model is capable of automatically decoupling global and local representations of images in an entirely unsupervised setting . The proposed model utilizes the variational auto-encoding framework to learn a (low-dimensional) vector of latent variables to capture the global information of an image, which is fed as a conditional input to a flow-based invertible decoder with architecture borrowed from style transfer literature . Experimental results on standard image benchmarks demonstrate the effectiveness of our model in terms of density estimation, image generation and unsuper supervised representation learning . Importantly, this work demonstrates that with only architectural inductive biases, a generative . model with a plain log-likelihood objective is capable

Links: PDF - Abstract

Code :


Keywords : model - generative - global - image - representations -

Leave a Reply

Your email address will not be published. Required fields are marked *