Towards Realistic Data Generation for Real-World Super-Resolution
This addresses the data generation bottleneck for real-world super-resolution, offering a practical solution for researchers and practitioners, though it is incremental as it builds on existing degradation and diffusion methods.
The paper tackles the problem of generating realistic training data for real-world image super-resolution, which suffers from poor generalization due to mismatched data, and introduces RealDGen, an unsupervised framework that creates large-scale, high-quality paired data, significantly improving SR model performance on benchmarks.
Existing image super-resolution (SR) techniques often fail to generalize effectively in complex real-world settings due to the significant divergence between training data and practical scenarios. To address this challenge, previous efforts have either manually simulated intricate physical-based degradations or utilized learning-based techniques, yet these approaches remain inadequate for producing large-scale, realistic, and diverse data simultaneously. In this paper, we introduce a novel Realistic Decoupled Data Generator (RealDGen), an unsupervised learning data generation framework designed for real-world super-resolution. We meticulously develop content and degradation extraction strategies, which are integrated into a novel content-degradation decoupled diffusion model to create realistic low-resolution images from unpaired real LR and HR images. Extensive experiments demonstrate that RealDGen excels in generating large-scale, high-quality paired data that mirrors real-world degradations, significantly advancing the performance of popular SR models on various real-world benchmarks.