Component Attention Guided Face Super-Resolution Network: CAGFace

19 Oct 2019Ratheesh KalarotTao LiFatih Porikli

To make the best use of the underlying structure of faces, the collective information through face datasets and the intermediate estimates during the upsampling process, here we introduce a fully convolutional multi-stage neural network for 4$\times$ super-resolution for face images. We implicitly impose facial component-wise attention maps using a segmentation network to allow our network to focus on face-inherent patterns... (read more)

PDF Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT LEADERBOARD
Image Super-Resolution FFHQ 1024 x 1024 - 4x upscaling CAGFace FID 12.4 # 1
MS-SSIM 0.971 # 1
PSNR 34.1 # 1
SSIM 0.906 # 1
Image Super-Resolution FFHQ 256 x 256 - 4x upscaling CAGFace FID 74.43 # 1
MS-SSIM 0.958 # 1
PSNR 27.42 # 1
SSIM 0.816 # 1