A Fully Progressive Approach to Single-Image Super-Resolution

Citations Over TimeTop 1% of 2018 papers

Abstract

Recent deep learning approaches to single image superresolution have achieved impressive results in terms of traditional error measures and perceptual quality. However, in each case it remains challenging to achieve high quality results for large upsampling factors. To this end, we propose a method (ProSR) that is progressive both in architecture and training: the network upsamples an image in intermediate steps, while the learning process is organized from easy to hard, as is done in curriculum learning. To obtain more photorealistic results, we design a generative adversarial network (GAN), named ProGanSR, that follows the same progressive multi-scale design principle. This not only allows to scale well to high upsampling factors (e.g., 8×) but constitutes a principled multi-scale approach that increases the reconstruction quality for all upsampling factors simultaneously. In particular ProSR ranks 2nd in terms of SSIM and 4th in terms of PSNR in the NTIRE2018 SISR challenge [35]. Compared to the top-ranking team, our model is marginally lower, but runs 5 times faster.

Related Papers

• → Bilateral Upsampling Network for Single Image Super-Resolution With Arbitrary Scaling Factors(2021)10 cited
• → Downsampling dependent upsampling of images(2003)32 cited
• → Fast algorithm of arbitrary factor subpixel downsampling based on frequency analysis(2014)1 cited
• → Noisy image upsampling based on improved bilateral filtering combined with edge extraction(2012)
• → Maximal level estimation and unbalance reduction for graph signal downsampling(2016)