Normalizing Flow as a Flexible Fidelity Objective for Photo-Realistic Super-resolution

TL;DR

This paper introduces a flow-based fidelity objective for photo-realistic super-resolution, improving visual quality and consistency over traditional pixel-wise losses by leveraging the flexibility of deep normalizing flows.

Contribution

It reinterprets the L_1 loss as a simple flow and proposes using deeper flows as a more flexible fidelity measure, enhancing super-resolution results.

Findings

Outperforms state-of-the-art methods in user studies

Deeper flows yield better visual quality

Flexibility of flows improves consistency with adversarial training

Abstract

Super-resolution is an ill-posed problem, where a ground-truth high-resolution image represents only one possibility in the space of plausible solutions. Yet, the dominant paradigm is to employ pixel-wise losses, such as L_1, which drive the prediction towards a blurry average. This leads to fundamentally conflicting objectives when combined with adversarial losses, which degrades the final quality. We address this issue by revisiting the L_1 loss and show that it corresponds to a one-layer conditional flow. Inspired by this relation, we explore general flows as a fidelity-based alternative to the L_1 objective. We demonstrate that the flexibility of deeper flows leads to better visual quality and consistency when combined with adversarial losses. We conduct extensive user studies for three datasets and scale factors, where our approach is shown to outperform state-of-the-art methods for photo-realistic super-resolution. Code and trained models will be available at: git.io/AdFlow