Super-resolution Using Constrained Deep Texture Synthesis
This addresses the challenge of enhancing image resolution for applications like photography and computer vision, though it appears incremental as it builds on existing deep texture synthesis methods.
The paper tackles the problem of oversmoothed outputs in single image super-resolution by using deep learning-based texture synthesis to transfer high-frequency details, improving visual quality across various natural images.
Hallucinating high frequency image details in single image super-resolution is a challenging task. Traditional super-resolution methods tend to produce oversmoothed output images due to the ambiguity in mapping between low and high resolution patches. We build on recent success in deep learning based texture synthesis and show that this rich feature space can facilitate successful transfer and synthesis of high frequency image details to improve the visual quality of super-resolution results on a wide variety of natural textures and images.