Super-Resolution through StyleGAN Regularized Latent Search: A Realism-Fidelity Trade-off
This addresses the problem of producing realistic and accurate super-resolved images for computer vision applications, but it is incremental as it builds on existing latent search methods.
The paper tackles super-resolution by searching the latent space of a StyleGAN, introducing a regularizer to keep images within the original manifold and expanding the image prior for better reconstruction, achieving realistic high-quality images with good fidelity-realism trade-off for various magnification factors.
This paper addresses the problem of super-resolution: constructing a highly resolved (HR) image from a low resolved (LR) one. Recent unsupervised approaches search the latent space of a StyleGAN pre-trained on HR images, for the image that best downscales to the input LR image. However, they tend to produce out-of-domain images and fail to accurately reconstruct HR images that are far from the original domain. Our contribution is twofold. Firstly, we introduce a new regularizer to constrain the search in the latent space, ensuring that the inverted code lies in the original image manifold. Secondly, we further enhanced the reconstruction through expanding the image prior around the optimal latent code. Our results show that the proposed approach recovers realistic high-quality images for large magnification factors. Furthermore, for low magnification factors, it can still reconstruct details that the generator could not have produced otherwise. Altogether, our approach achieves a good trade-off between fidelity and realism for the super-resolution task.