SRDiff: Single Image Super-Resolution with Diffusion Probabilistic Models

TL;DR

SRDiff introduces the first diffusion-based approach for single image super-resolution, producing diverse, detailed, and realistic high-resolution images from low-resolution inputs with improved training efficiency.

Contribution

The paper presents a novel diffusion probabilistic model for SISR that overcomes over-smoothing and mode collapse issues, enabling diverse and high-quality super-resolution results.

Findings

Achieves state-of-the-art performance on CelebA and DIV2K datasets.

Generates diverse super-resolution images with rich details.

Supports flexible image manipulation such as latent space interpolation.

Abstract

Single image super-resolution (SISR) aims to reconstruct high-resolution (HR) images from the given low-resolution (LR) ones, which is an ill-posed problem because one LR image corresponds to multiple HR images. Recently, learning-based SISR methods have greatly outperformed traditional ones, while suffering from over-smoothing, mode collapse or large model footprint issues for PSNR-oriented, GAN-driven and flow-based methods respectively. To solve these problems, we propose a novel single image super-resolution diffusion probabilistic model (SRDiff), which is the first diffusion-based model for SISR. SRDiff is optimized with a variant of the variational bound on the data likelihood and can provide diverse and realistic SR predictions by gradually transforming the Gaussian noise into a super-resolution (SR) image conditioned on an LR input through a Markov chain. In addition, we introduce residual prediction to the whole framework to speed up convergence. Our extensive experiments on facial and general benchmarks (CelebA and DIV2K datasets) show that 1) SRDiff can generate diverse SR results in rich details with state-of-the-art performance, given only one LR input; 2) SRDiff is easy to train with a small footprint; and 3) SRDiff can perform flexible image manipulation including latent space interpolation and content fusion.