Domain Transfer in Latent Space (DTLS) Wins on Image Super-Resolution -- a Non-Denoising Model
Authors:
Chun-Chuen Hui,
Wan-Chi Siu,
Ngai-Fong Law
Abstract:
Large scale image super-resolution is a challenging computer vision task, since vast information is missing in a highly degraded image, say for example forscale x16 super-resolution. Diffusion models are used successfully in recent years in extreme super-resolution applications, in which Gaussian noise is used as a means to form a latent photo-realistic space, and acts as a link between the space…
▽ More
Large scale image super-resolution is a challenging computer vision task, since vast information is missing in a highly degraded image, say for example forscale x16 super-resolution. Diffusion models are used successfully in recent years in extreme super-resolution applications, in which Gaussian noise is used as a means to form a latent photo-realistic space, and acts as a link between the space of latent vectors and the latent photo-realistic space. There are quite a few sophisticated mathematical derivations on mapping the statistics of Gaussian noises making Diffusion Models successful. In this paper we propose a simple approach which gets away from using Gaussian noise but adopts some basic structures of diffusion models for efficient image super-resolution. Essentially, we propose a DNN to perform domain transfer between neighbor domains, which can learn the differences in statistical properties to facilitate gradual interpolation with results of reasonable quality. Further quality improvement is achieved by conditioning the domain transfer with reference to the input LR image. Experimental results show that our method outperforms not only state-of-the-art large scale super resolution models, but also the current diffusion models for image super-resolution. The approach can readily be extended to other image-to-image tasks, such as image enlightening, inpainting, denoising, etc.
△ Less
Submitted 21 December, 2023; v1 submitted 4 November, 2023;
originally announced November 2023.
