Progressive Face Super-Resolution with Non-Parametric Facial Prior Enhancement

Abstract

The main challenge of face super-resolution is to overcome facial distortions in an upscaling process. Recent works have utilized facial priors such as facial landmarks and component maps to generate a precise super-resolved image. However, the facial priors are estimated from the ground-truth and deep neural networks. Thus, recent works based on the facial priors are not only limited to specific datasets including the groundtruth, but also need sub-networks to extract facial priors. To solve these problems, we propose a progressive face superresolution network with non-parametric facial prior enhancement, called as NPFNet, which extracts and highlights facial components without any tricks, such as the ground-truth and deep neural networks. The self-enhancement module facilitates our network to fully utilize facial distinct features to enhance super-resolved images with a parameter-free operation. Extensive experiments on CelebA and VGGFace2 demonstrate that the proposed method outperforms state-of-the-art face super-resolution methods in terms of visual quality and quantitative measurements.

Towards Real-World Blind Face Restoration with Generative Facial Prior

“…Based on general face hallucination [5,31,68,71], two typical face-specific priors: geometry priors and reference priors, are incorporated to further improve the performance. The geometry priors include facial landmarks [9,38,78], face parsing maps [60,6,9] and facial component heatmaps [70]. However, 1) those priors require estimations from low-quality inputs and inevitably degrades in real-world scenarios.…”

GAN Prior Embedded Network for Blind Face Restoration in the Wild

“…Nishiyama et al [36] proposed to improve the recognition performance of blurry faces by using a pre-defined set of blur kernels to restore them. With the unprecedented success of DNNs in solving image restoration tasks such as denoising [13], deblur- ring [24,40], inpainting [48,31] and image super-resolution [25,30], many DNN based face image restoration methods have also been proposed [7,23,33], which advance the traditional methods by a large margin. Considering the fact that facial images have specific structures, it is interesting to investigate whether we can restore a clear face image from severely degraded ones without knowing the degradation model.…”

GLEAN: Generative Latent Bank for Large-Factor Image Super-Resolution

“…SR methods specialized on large magnification factors are typically dedicated to the human face category as one could exploit the strong structural prior of faces. Facial priors including facial attributes [22], facial landmarks [18,26], and identity [10] have been studied. Our work goes beyond previous works and pushes the limit to 64× and generalizes to more categories.…”