Self-Organized Residual Blocks for Image Super-Resolution

TL;DR

This paper introduces self-organized operational residual blocks for image super-resolution, combining residual and self-organized neural network concepts to improve performance in PSNR and perceptual quality.

Contribution

It proposes a novel hybrid network architecture with self-organized residual blocks that enhance non-linearity and performance in image super-resolution tasks.

Findings

Improved PSNR and perceptual metrics over baseline models

Hybrid architectures balance non-linearity benefits and parameter efficiency

Self-organized residual blocks outperform traditional convolutional blocks

Abstract

It has become a standard practice to use the convolutional networks (ConvNet) with RELU non-linearity in image restoration and super-resolution (SR). Although the universal approximation theorem states that a multi-layer neural network can approximate any non-linear function with the desired precision, it does not reveal the best network architecture to do so. Recently, operational neural networks (ONNs) that choose the best non-linearity from a set of alternatives, and their "self-organized" variants (Self-ONN) that approximate any non-linearity via Taylor series have been proposed to address the well-known limitations and drawbacks of conventional ConvNets such as network homogeneity using only the McCulloch-Pitts neuron model. In this paper, we propose the concept of self-organized operational residual (SOR) blocks, and present hybrid network architectures combining regular residual and SOR blocks to strike a balance between the benefits of stronger non-linearity and the overall number of parameters. The experimental results demonstrate that the~proposed architectures yield performance improvements in both PSNR and perceptual metrics.