A very lightweight and efficient image super-resolution network

作者：

Highlights：

• A very lightweight and efficient image super-resolution network has been proposed.

• A high-/low-frequency feature grouping fusion block has been proposed.

• A multi-way attention block has been proposed.

• A lightweight residual concatenation block has been proposed.

• A progressive interactive group convolution has been proposed.

摘要

•A very lightweight and efficient image super-resolution network has been proposed.•A high-/low-frequency feature grouping fusion block has been proposed.•A multi-way attention block has been proposed.•A lightweight residual concatenation block has been proposed.•A progressive interactive group convolution has been proposed.

论文关键词：Image super-resolution,Lightweight,Group convolution,Attention mechanism,Feature fusion

论文评审过程：Received 26 January 2022, Revised 20 September 2022, Accepted 21 September 2022, Available online 28 September 2022, Version of Record 7 October 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118898