论文标题
收集：大型图像超分辨率的生成潜在银行
GLEAN: Generative Latent Bank for Large-Factor Image Super-Resolution
论文作者
论文摘要
我们表明，预训练的生成对抗网络（GAN），例如Stylegan，可以用作潜在银行，以提高大型图像超分辨率（SR）的恢复质量。尽管大多数现有的SR方法试图通过以对抗性损失学习来生成逼真的纹理，但我们的方法，即生成潜伏银行（GLEAN），通过直接利用在预先训练的GAN中封装的丰富而多样的先验来超越现有实践。但是，与需要在运行时需要昂贵图像特定优化的普遍的GAN反演方法不同，我们的方法只需要一个正向通行证即可生成高尺度的图像。可以轻松地将GLEAN合并到具有多分辨率Skip连接的简单编码器银行decoder架构中。切换银行允许该方法处理来自不同类别的图像，例如猫，建筑物，人脸和汽车。与现有方法相比，Glean升级的图像在忠诚度和质地忠诚方面显示出明显的改善。
We show that pre-trained Generative Adversarial Networks (GANs), e.g., StyleGAN, can be used as a latent bank to improve the restoration quality of large-factor image super-resolution (SR). While most existing SR approaches attempt to generate realistic textures through learning with adversarial loss, our method, Generative LatEnt bANk (GLEAN), goes beyond existing practices by directly leveraging rich and diverse priors encapsulated in a pre-trained GAN. But unlike prevalent GAN inversion methods that require expensive image-specific optimization at runtime, our approach only needs a single forward pass to generate the upscaled image. GLEAN can be easily incorporated in a simple encoder-bank-decoder architecture with multi-resolution skip connections. Switching the bank allows the method to deal with images from diverse categories, e.g., cat, building, human face, and car. Images upscaled by GLEAN show clear improvements in terms of fidelity and texture faithfulness in comparison to existing methods.
