Two Heads Better than One: Dual Degradation Representation for Blind Super-Resolution
This addresses the performance decline in super-resolution for real-world images with unknown degradation, offering a domain-specific improvement for computer vision applications.
The paper tackles the blind super-resolution problem where actual degradation deviates from known assumptions, by proposing a Dual Branch Degradation Extractor Network that predicts unsupervised blur and noise embeddings to adapt the SR network, achieving state-of-the-art performance on benchmarks.
Previous methods have demonstrated remarkable performance in single image super-resolution (SISR) tasks with known and fixed degradation (e.g., bicubic downsampling). However, when the actual degradation deviates from these assumptions, these methods may experience significant declines in performance. In this paper, we propose a Dual Branch Degradation Extractor Network to address the blind SR problem. While some blind SR methods assume noise-free degradation and others do not explicitly consider the presence of noise in the degradation model, our approach predicts two unsupervised degradation embeddings that represent blurry and noisy information. The SR network can then be adapted to blur embedding and noise embedding in distinct ways. Furthermore, we treat the degradation extractor as a regularizer to capitalize on differences between SR and HR images. Extensive experiments on several benchmarks demonstrate our method achieves SOTA performance in the blind SR problem.