Efficient Single Image Super Resolution using Enhanced Learned Group Convolutions

TL;DR

The paper introduces SRCondenseNet, a computationally efficient CNN-based method for single-image super-resolution that outperforms existing models in accuracy while reducing computational costs.

Contribution

It presents a novel super-resolution approach using refined group convolutions and a deconvolutional reconstruction network, optimizing for lower computation without sacrificing performance.

Findings

Outperforms state-of-the-art methods in accuracy

Reduces computational requirements significantly

Effective use of group convolutions and bicubic input addition

Abstract

Convolutional Neural Networks (CNNs) have demonstrated great results for the single-image super-resolution (SISR) problem. Currently, most CNN algorithms promote deep and computationally expensive models to solve SISR. However, we propose a novel SISR method that uses relatively less number of computations. On training, we get group convolutions that have unused connections removed. We have refined this system specifically for the task at hand by removing unnecessary modules from original CondenseNet. Further, a reconstruction network consisting of deconvolutional layers has been used in order to upscale to high resolution. All these steps significantly reduce the number of computations required at testing time. Along with this, bicubic upsampled input is added to the network output for easier learning. Our model is named SRCondenseNet. We evaluate the method using various benchmark datasets and show that it performs favourably against the state-of-the-art methods in terms of both accuracy and number of computations required.