SSNeT: A Lightweight Residual Network for Single Image Super Resolution

View at publisher Summarise with VIRAT
Authors: P G Prathibha, P. S. Tamizharasan
Cited by: 0
References: 33

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

In recent years, Convolutional Neural Networks enabled notable advancements in the domain of single-image super-resolution. However, the high computational demands and resource requirements of most of the previous research have hindered their feasibility for deployment in real-time applications. In this article we introduce a single image super-resolution model that explores lightweight residual block and attention mechanism. We designed a lightweight residual block based on Sandblock architecture, which incorporates an inverted residual block, resulting in an hourglass-shaped structure. An attention module is integrated inorder to enhance the features extracted by the residual block. The modules provide extensive and refined feature extraction with very less number of parameters The proposed approach yields an improvement of 0.48db from the second-best performing model on Set14 dataset and enhanced performance on other datasets

Fields of Study & Subjects

Advanced Image Processing Techniques Advanced Neural Network Applications Image and Signal Denoising Methods

Publication details

Journal / source
International Conference on Consumer Electronics
Pages
1–6
Publication date
3 February 2026
Language
EN