Deep Back-Projection Networks for Single Image Super-resolution

TL;DR

This paper introduces Deep Back-Projection Networks (DBPN), a novel super-resolution architecture that uses iterative error feedback through up- and down-sampling units, achieving state-of-the-art results especially for large scaling factors.

Contribution

The paper presents a new deep network architecture with iterative back-projection units that improve super-resolution performance over previous feed-forward models.

Findings

Achieved state-of-the-art results on multiple datasets.

Excelled particularly at 8x super-resolution scaling.

Won two major super-resolution challenges (NTIRE2018, PIRM2018).

Abstract

Previous feed-forward architectures of recently proposed deep super-resolution networks learn the features of low-resolution inputs and the non-linear mapping from those to a high-resolution output. However, this approach does not fully address the mutual dependencies of low- and high-resolution images. We propose Deep Back-Projection Networks (DBPN), the winner of two image super-resolution challenges (NTIRE2018 and PIRM2018), that exploit iterative up- and down-sampling layers. These layers are formed as a unit providing an error feedback mechanism for projection errors. We construct mutually-connected up- and down-sampling units each of which represents different types of low- and high-resolution components. We also show that extending this idea to demonstrate a new insight towards more efficient network design substantially, such as parameter sharing on the projection module and transition layer on projection step. The experimental results yield superior results and in particular establishing new state-of-the-art results across multiple data sets, especially for large scaling factors such as 8x.