Explicitly Minimizing the Blur Error of Variational Autoencoders

Gustav Bredell, Kyriakos Flouris, Krishna Chaitanya, Ertunc Erdil, Ender Konukoglu

Keywords: Variational Autoencoders, Generative Modelling, Blur

TL;DR: We propose a new reconstruction term for VAEs that explicitly focuses on minimizing the blur of generated/reconstructed images while still optimizing the ELBO.

Abstract:

Variational autoencoders (VAEs) are powerful generative modelling methods, however they suffer from blurry generated samples and reconstructions compared to the images they have been trained on. Significant research effort has been spent to increase the generative capabilities by creating more flexible models but often flexibility comes at the cost of higher complexity and computational cost. Several works have focused on altering the reconstruction term of the evidence lower bound (ELBO), however, often at the expense of losing the mathematical link to maximizing the likelihood of the samples under the modeled distribution. Here we propose a new formulation of the reconstruction term for the VAE that specifically penalizes the generation of blurry images while at the same time still maximizing the ELBO under the modeled distribution.
We show the potential of the proposed loss on three different data sets, where it outperforms several recently proposed reconstruction losses for VAEs.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Generative models