Image Processing Using Multi-Code GAN Prior

CVPR 2020  ·  Jinjin Gu, Yujun Shen, Bolei Zhou ·

Despite the success of Generative Adversarial Networks (GANs) in image synthesis, applying trained GAN models to real image processing remains challenging. Previous methods typically invert a target image back to the latent space either by back-propagation or by learning an additional encoder... However, the reconstructions from both of the methods are far from ideal. In this work, we propose a novel approach, called mGANprior, to incorporate the well-trained GANs as effective prior to a variety of image processing tasks. In particular, we employ multiple latent codes to generate multiple feature maps at some intermediate layer of the generator, then compose them with adaptive channel importance to recover the input image. Such an over-parameterization of the latent space significantly improves the image reconstruction quality, outperforming existing competitors. The resulting high-fidelity image reconstruction enables the trained GAN models as prior to many real-world applications, such as image colorization, super-resolution, image inpainting, and semantic manipulation. We further analyze the properties of the layer-wise representation learned by GAN models and shed light on what knowledge each layer is capable of representing. read more

PDF Abstract CVPR 2020 PDF CVPR 2020 Abstract

Datasets


Results from Other Papers


Task Dataset Model Metric Name Metric Value Rank Source Paper Compare
Blind Face Restoration CelebA-Test mGANprior LPIPS 45.84 # 4
FID 82.27 # 5
NIQE 6.422 # 5
Deg. 55.45 # 5
PSNR 24.30 # 5
SSIM 0.6758 # 3

Methods