Sub-Pixel Back-Projection Network For Lightweight Single Image Super-Resolution

3 Aug 2020  ·  Supratik Banerjee, Cagri Ozcinar, Aakanksha Rana, Aljosa Smolic, Michael Manzke ·

Convolutional neural network (CNN)-based methods have achieved great success for single-image superresolution (SISR). However, most models attempt to improve reconstruction accuracy while increasing the requirement of number of model parameters. To tackle this problem, in this paper, we study reducing the number of parameters and computational cost of CNN-based SISR methods while maintaining the accuracy of super-resolution reconstruction performance. To this end, we introduce a novel network architecture for SISR, which strikes a good trade-off between reconstruction quality and low computational complexity. Specifically, we propose an iterative back-projection architecture using sub-pixel convolution instead of deconvolution layers. We evaluate the performance of computational and reconstruction accuracy for our proposed model with extensive quantitative and qualitative evaluations. Experimental results reveal that our proposed method uses fewer parameters and reduces the computational cost while maintaining reconstruction accuracy against state-of-the-art SISR methods over well-known four SR benchmark datasets. Code is available at "https://github.com/supratikbanerjee/SubPixel-BackProjection_SuperResolution".

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Image Super-Resolution BSDS100 - 2x upscaling SPBP-L+ PSNR 32.21 # 2
SSIM 0.9001 # 2
Image Super-Resolution Set14 - 2x upscaling SPBP-L+ PSNR 33.62 # 16
SSIM 0.9178 # 12
Image Super-Resolution Set5 - 2x upscaling SPBP-L+ PSNR 38.05 # 16
SSIM 0.9606 # 11
Image Super-Resolution Urban100 - 2x upscaling SPBP-L+ PSNR 32.07 # 17
SSIM 0.9277 # 11

Methods