Towards Real-World Blind Face Restoration with Generative Facial Prior

CVPR 2021  ·  Xintao Wang, Yu Li, Honglun Zhang, Ying Shan ·

Blind face restoration usually relies on facial priors, such as facial geometry prior or reference prior, to restore realistic and faithful details. However, very low-quality inputs cannot offer accurate geometric prior while high-quality references are inaccessible, limiting the applicability in real-world scenarios. In this work, we propose GFP-GAN that leverages rich and diverse priors encapsulated in a pretrained face GAN for blind face restoration. This Generative Facial Prior (GFP) is incorporated into the face restoration process via novel channel-split spatial feature transform layers, which allow our method to achieve a good balance of realness and fidelity. Thanks to the powerful generative facial prior and delicate designs, our GFP-GAN could jointly restore facial details and enhance colors with just a single forward pass, while GAN inversion methods require expensive image-specific optimization at inference. Extensive experiments show that our method achieves superior performance to prior art on both synthetic and real-world datasets.

PDF Abstract CVPR 2021 PDF CVPR 2021 Abstract

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Blind Face Restoration CelebA-Test GFP-GAN LPIPS 36.46 # 1
FID 42.62 # 1
NIQE 4.077 # 1
Deg. 34.60 # 1
PSNR 25.08 # 2
SSIM 0.6777 # 2
Video Super-Resolution MSU Video Super Resolution Benchmark: Detail Restoration GFPGAN Subjective score 2.686 # 30
ERQAv1.0 0.538 # 30
QRCRv1.0 0 # 21
SSIM 0.745 # 32
PSNR 24.195 # 31
FPS 1.562 # 10
1 - LPIPS 0.793 # 27

Methods