# Advanced Super-Resolution using Lossless Pooling Convolutional Networks

## TL;DR

This paper introduces a novel super-resolution method that leverages auxiliary self-replicas of the input image and lossless pooling layers within a neural network to improve upscaling quality beyond traditional models.

## Contribution

It proposes a new deep learning architecture that incorporates artificially created self-replicas and lossless pooling layers for enhanced image super-resolution.

## Key findings

- Significant improvements in super-resolution quality confirmed by extensive evaluations.
- The method effectively exploits high correlation between multiple image instances.
- Outperforms existing super-resolution approaches in various metrics.

## Abstract

In this paper, we present a novel deep learning-based approach for still image super-resolution, that unlike the mainstream models does not rely solely on the input low resolution image for high quality upsampling, and takes advantage of a set of artificially created auxiliary self-replicas of the input image that are incorporated in the neural network to create an enhanced and accurate upscaling scheme. Inclusion of the proposed lossless pooling layers, and the fusion of the input self-replicas enable the model to exploit the high correlation between multiple instances of the same content, and eventually result in significant improvements in the quality of the super-resolution, which is confirmed by extensive evaluations.