Image Super-Resolution with Cross-Scale Non-Local Attention and Exhaustive Self-Exemplars Mining

Deep convolution-based single image super-resolution (SISR) networks embrace the benefits of learning from large-scale external image resources for local recovery, yet most existing works have ignored the long-range feature-wise similarities in natural images. Some recent works have successfully leveraged this intrinsic feature correlation by exploring non-local attention modules... (read more)

PDF Abstract CVPR 2020 PDF CVPR 2020 Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Image Super-Resolution BSD100 - 2x upscaling CSNLN PSNR 32.4 # 4
SSIM 0.9024 # 3
Image Super-Resolution BSD100 - 4x upscaling CSNLN PSNR 27.8 # 7
SSIM 0.7439 # 11
Image Super-Resolution Manga109 - 2x upscaling CSNLN PSNR 39.37 # 3
SSIM 0.9785 # 4
Image Super-Resolution Manga109 - 4x upscaling CSNLN PSNR 31.43 # 8
SSIM 0.9201 # 5
Image Super-Resolution Set14 - 2x upscaling CSNLN SSIM 0.9223 # 4
Image Super-Resolution Set14 - 4x upscaling CSNLN PSNR 28.95 # 7
SSIM 0.7888 # 15
Image Super-Resolution Set5 - 2x upscaling CSNLN PSNR 38.28 # 4
SSIM 0.9616 # 3
Image Super-Resolution Set5 - 4x upscaling CSNLN PSNR 32.68 # 8
SSIM 0.9004 # 9
Image Super-Resolution Urban100 - 2x upscaling CSNLN PSNR 33.25 # 4
SSIM 0.9386 # 3
Image Super-Resolution Urban100 - 4x upscaling CSNLN PSNR 27.22 # 4
SSIM 0.8168 # 4

Methods used in the Paper


METHOD TYPE
🤖 No Methods Found Help the community by adding them if they're not listed; e.g. Deep Residual Learning for Image Recognition uses ResNet