Image Super-Resolution via Iterative Refinement

TL;DR

SR3 introduces a novel iterative denoising diffusion approach for image super-resolution, achieving high-quality, photo-realistic results and outperforming GAN-based methods in human evaluations and cascaded generation tasks.

Contribution

The paper adapts denoising diffusion models for super-resolution, demonstrating superior performance and realism compared to existing GAN-based methods.

Findings

Achieves near 50% fool rate on face super-resolution, indicating high realism.

Outperforms GANs in human evaluations for super-resolution quality.

Provides competitive FID score of 11.3 in cascaded image generation.

Abstract

We present SR3, an approach to image Super-Resolution via Repeated Refinement. SR3 adapts denoising diffusion probabilistic models to conditional image generation and performs super-resolution through a stochastic denoising process. Inference starts with pure Gaussian noise and iteratively refines the noisy output using a U-Net model trained on denoising at various noise levels. SR3 exhibits strong performance on super-resolution tasks at different magnification factors, on faces and natural images. We conduct human evaluation on a standard 8X face super-resolution task on CelebA-HQ, comparing with SOTA GAN methods. SR3 achieves a fool rate close to 50%, suggesting photo-realistic outputs, while GANs do not exceed a fool rate of 34%. We further show the effectiveness of SR3 in cascaded image generation, where generative models are chained with super-resolution models, yielding a competitive FID score of 11.3 on ImageNet.