Component Attention Guided Face Super-Resolution Network: CAGFace

19 Oct 2019  ·  Ratheesh Kalarot, Tao Li, Fatih 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. Each stage of our network is composed of a stem layer, a residual backbone, and spatial upsampling layers. We recurrently apply stages to reconstruct an intermediate image, and then reuse its space-to-depth converted versions to bootstrap and enhance image quality progressively. Our experiments show that our face super-resolution method achieves quantitatively superior and perceptually pleasing results in comparison to state of the art.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Image Super-Resolution FFHQ 1024 x 1024 - 4x upscaling CAGFace FID 12.4 # 2
MS-SSIM 0.971 # 2
PSNR 34.1 # 1
SSIM 0.906 # 1
Image Super-Resolution FFHQ 256 x 256 - 4x upscaling CAGFace FID 74.43 # 2
MS-SSIM 0.958 # 2
PSNR 27.42 # 2
SSIM 0.816 # 1


No methods listed for this paper. Add relevant methods here