Researchers at Tencent Propose GFP-GAN that Leverages Rich and Diverse Priors Encapsulated in a Pretrained Face GAN for Blind Face Restoration

The goal of blind face restoration is to recover high-quality images of human faces from their low-quality counterparts that have been degraded for an unknown reason. Some degradation causes could be noise, blur, low-resolution, and compression artifacts. In this work, researchers from the Applied Research Center of the Tencent company propose [GFP-GAN](https://github.com/TencentARC/GFPGAN), a Generative Facial Prior GAN for real-world blind face restoration. As it is possible to see in Figure 1, the images restored through GFP-GAN reach higher realness and …

computervision face gan researchers tencent