Image Superresolution using Scale-Recurrent Dense Network

TL;DR

This paper introduces a scale recurrent dense network with residual dense blocks for image super-resolution, achieving high performance with fewer parameters by combining dense connections, residual links, and GAN-based training.

Contribution

It proposes a novel scale recurrent architecture with multi-residual dense blocks and GAN-based loss functions, improving efficiency and perceptual quality in image super-resolution.

Findings

Competitive performance at high scale factors

Fewer parameters than state-of-the-art methods

Enhanced perceptual quality through GAN and VGG losses

Abstract

Recent advances in the design of convolutional neural network (CNN) have yielded significant improvements in the performance of image super-resolution (SR). The boost in performance can be attributed to the presence of residual or dense connections within the intermediate layers of these networks. The efficient combination of such connections can reduce the number of parameters drastically while maintaining the restoration quality. In this paper, we propose a scale recurrent SR architecture built upon units containing series of dense connections within a residual block (Residual Dense Blocks (RDBs)) that allow extraction of abundant local features from the image. Our scale recurrent design delivers competitive performance for higher scale factors while being parametrically more efficient as compared to current state-of-the-art approaches. To further improve the performance of our network, we employ multiple residual connections in intermediate layers (referred to as Multi-Residual Dense Blocks), which improves gradient propagation in existing layers. Recent works have discovered that conventional loss functions can guide a network to produce results which have high PSNRs but are perceptually inferior. We mitigate this issue by utilizing a Generative Adversarial Network (GAN) based framework and deep feature (VGG) losses to train our network. We experimentally demonstrate that different weighted combinations of the VGG loss and the adversarial loss enable our network outputs to traverse along the perception-distortion curve. The proposed networks perform favorably against existing methods, both perceptually and objectively (PSNR-based) with fewer parameters.