Abstract:Recent research on single-image super-resolution (SISR) using deep convolutional neural networks has made a breakthrough and achieved tremendous performance. Despite their significant progress, numerous convolutional neural networks (CNN) are limited in practical applications, owing to the requirement of the heavy computational cost of the model. This paper proposes a multi-path network for SISR, known as multi-path deep CNN with residual inception network for single image super-resolution. In detail, a residual/ResNet block with an Inception block supports the main framework of the entire network architecture. In addition, remove the batch normalization layer from the residual network (ResNet) block and max-pooling layer from the Inception block to further reduce the number of parameters to preventing the over-fitting problem during the training. Moreover, a conventional rectified linear unit (ReLU) is replaced with Leaky ReLU activation function to speed up the training process. Specifically, we propose a novel two upscale module, which adopts three paths to upscale the features by jointly using deconvolution and upsampling layers, instead of using single deconvolution layer or upsampling layer alone. The extensive experimental results on image super-resolution (SR) using five publicly available test datasets, which show that the proposed model not only attains the higher score of peak signal-to-noise ratio/structural similarity index matrix (PSNR/SSIM) but also enables faster and more efficient calculations against the existing image SR methods. For instance, we improved our method in terms of overall PSNR on the SET5 dataset with challenging upscale factor 8× as 1.88 dB over the baseline bicubic method and reduced computational cost in terms of number of parameters 62% by deeply-recursive convolutional neural network (DRCN) method.

An efficient multi-scaled super resolution framework for image enhancement

An Efficient Multi-scale Method for Single Image Super Resolution.

Image Super-Resolution Via a Novel Cascaded Convolutional Neural Network Framework.

Multi-scale Feature Selection Network for Lightweight Image Super-Resolution

Compressed Multi-Scale Feature Fusion Network for Single Image Super-Resolution.

Multi-Path Deep CNN with Residual Inception Network for Single Image Super-Resolution

Lightweight multi-scale residual networks with attention for image super-resolution

Learning Depth Super-Resolution by Using Multi-Scale Convolutional Neural Network.

Multi-Scale Residual Network For Image Super-Resolution

Lightweight Image Super-Resolution Via Weighted Multi-Scale Residual Network

Efficient Self-Calibrated and Hierarchical Refinement Network for Lightweight Super-Resolution

A Multi-Scale Recursive Attention Feature Fusion Network for Image Super-Resolution Reconstruction Algorithm

An Image Super-Resolution Network Based on Multi-Scale Convolution Fusion

End-to-End Image Super-Resolution via Deep and Shallow Convolutional Networks

The Multi-scale Fast Network for Image Super-resolution

Image super-resolution via dynamic network

Image Super-Resolution via Deep Feature Recalibration Network.

Efficient local cascading residual network for real-time single image super-resolution

Image super-resolution via enhanced multi-scale residual network

Multi-scale skip-connection network for image super-resolution

Single image super-resolution via deep progressive multi-scale fusion networks