Towards Real-World Blind Face Restoration with Generative Facial Prior
This work provides a more robust solution for real-world blind face restoration, particularly for scenarios with very low-quality input images, benefiting applications like old photo restoration or surveillance image enhancement.
The paper addresses the challenge of blind face restoration from very low-quality inputs where traditional facial priors are inaccurate or unavailable. The authors propose GFP-GAN, which utilizes a pretrained face GAN as a Generative Facial Prior to restore facial details and enhance colors in a single forward pass, outperforming prior methods on synthetic and real-world datasets.
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.