Lightweight Feature Fusion Network for Single Image Super-Resolution

15 Feb 2019  ·  Wenming Yang, Wei Wang, Xuechen Zhang, Shuifa Sun, Qingmin Liao ·

Single image super-resolution(SISR) has witnessed great progress as convolutional neural network(CNN) gets deeper and wider. However, enormous parameters hinder its application to real world problems. In this letter, We propose a lightweight feature fusion network (LFFN) that can fully explore multi-scale contextual information and greatly reduce network parameters while maximizing SISR results. LFFN is built on spindle blocks and a softmax feature fusion module (SFFM). Specifically, a spindle block is composed of a dimension extension unit, a feature exploration unit and a feature refinement unit. The dimension extension layer expands low dimension to high dimension and implicitly learns the feature maps which is suitable for the next unit. The feature exploration unit performs linear and nonlinear feature exploration aimed at different feature maps. The feature refinement layer is used to fuse and refine features. SFFM fuses the features from different modules in a self-adaptive learning manner with softmax function, making full use of hierarchical information with a small amount of parameter cost. Both qualitative and quantitative experiments on benchmark datasets show that LFFN achieves favorable performance against state-of-the-art methods with similar parameters.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Image Super-Resolution BSD100 - 2x upscaling LFFN-S PSNR 31.96 # 16
Image Super-Resolution BSD100 - 3x upscaling LFFN-S PSNR 28.91 # 11
Image Super-Resolution BSD100 - 4x upscaling LFFN-S PSNR 27.42 # 32
Image Super-Resolution Manga109 - 2x upscaling LFFN-S PSNR 37.93 # 12
SSIM 0.9746 # 10
Image Super-Resolution Manga109 - 3x upscaling LFFN-S PSNR 32.8 # 9
SSIM 0.9381 # 7
Image Super-Resolution Manga109 - 4x upscaling LFFN-S PSNR 29.76 # 21
SSIM 0.8979 # 21
Image Super-Resolution Set5 - 2x upscaling LFFN-S PSNR 37.66 # 18
SSIM 0.9585 # 12
Image Super-Resolution Set5 - 3x upscaling LFFN-S PSNR 34.04 # 12
SSIM 0.9233 # 7
Image Super-Resolution Set5 - 4x upscaling LFFN-S PSNR 31.79 # 36
SSIM 0.8886 # 37

Methods