Continuous Degradation Modeling via Latent Flow Matching for Real-World Super-Resolution

TL;DR

This paper presents a novel framework that synthesizes realistic low-resolution images from high-resolution images using latent flow matching, enabling better training data for real-world super-resolution models.

Contribution

It introduces a new method for generating authentic LR images from HR images via latent flow matching, improving training data quality for real-world SR tasks.

Findings

Synthetic LR images closely mimic real-world degradations.

SR models trained on our datasets outperform those trained on synthetic data.

Our approach enables arbitrary-scale super-resolution training.

Abstract

While deep learning-based super-resolution (SR) methods have shown impressive outcomes with synthetic degradation scenarios such as bicubic downsampling, they frequently struggle to perform well on real-world images that feature complex, nonlinear degradations like noise, blur, and compression artifacts. Recent efforts to address this issue have involved the painstaking compilation of real low-resolution (LR) and high-resolution (HR) image pairs, usually limited to several specific downscaling factors. To address these challenges, our work introduces a novel framework capable of synthesizing authentic LR images from a single HR image by leveraging the latent degradation space with flow matching. Our approach generates LR images with realistic artifacts at unseen degradation levels, which facilitates the creation of large-scale, real-world SR training datasets. Comprehensive quantitative and qualitative assessments verify that our synthetic LR images accurately replicate real-world degradations. Furthermore, both traditional and arbitrary-scale SR models trained using our datasets consistently yield much better HR outcomes.