MMSR: Multiple-Model Learned Image Super-Resolution Benefiting From Class-Specific Image Priors
This work addresses the issue of poor generalization in single SR models for diverse image content, offering a solution for applications requiring high-quality image upscaling across various domains.
The paper tackles the problem of image super-resolution (SR) by proposing a multiple-model approach that trains separate SR models for different image classes (e.g., text, texture) and uses a post-processing network to fuse their outputs, which significantly outperforms state-of-the-art generic SR models both quantitatively and visually.
Assuming a known degradation model, the performance of a learned image super-resolution (SR) model depends on how well the variety of image characteristics within the training set matches those in the test set. As a result, the performance of an SR model varies noticeably from image to image over a test set depending on whether characteristics of specific images are similar to those in the training set or not. Hence, in general, a single SR model cannot generalize well enough for all types of image content. In this work, we show that training multiple SR models for different classes of images (e.g., for text, texture, etc.) to exploit class-specific image priors and employing a post-processing network that learns how to best fuse the outputs produced by these multiple SR models surpasses the performance of state-of-the-art generic SR models. Experimental results clearly demonstrate that the proposed multiple-model SR (MMSR) approach significantly outperforms a single pre-trained state-of-the-art SR model both quantitatively and visually. It even exceeds the performance of the best single class-specific SR model trained on similar text or texture images.