NTIRE 2017 Challenge on Single Image Super-Resolution: Dataset and Study

This paper introduces a novel large dataset for example-based single image super-resolution and studies the state-of-the-art as emerged from the NTIRE 2017 challenge. The challenge is the first challenge of its kind, with 6 competitions, hundreds of participants and tens of proposed solutions. Our newly collected DIVerse 2K resolution image dataset (DIV2K) was employed by the challenge. In our study we compare the solutions from the challenge to a set of representative methods from the literature and evaluate them using diverse measures on our proposed DIV2K dataset. Moreover, we conduct a number of experiments and draw conclusions on several topics of interest. We conclude that the NTIRE 2017 challenge pushes the state-of-the-art in single-image super-resolution, reaching the best results to date on the popular Set5, Set14, B100, Urban100 datasets and on our newly proposed DIV2K.

1. Introduction

Example-based single image super-resolution (SR) aims at full restoration of rich details (high frequencies) in images based on prior examples under the form of low resolution (LR) and corresponding high resolution (HR) images. The loss of details / contents can be due to various degrading factors such as blur, decimation, noise or hardware limitations (e.g. camera sensors). SR is an ill-posed problem because for each LR image patch the number of corresponding HR image patches can be very large.