论文标题
在生成模型和推论自动编码器中使用结构相似性指数的理论见解
Theoretical Insights into the Use of Structural Similarity Index In Generative Models and Inferential Autoencoders
论文作者
论文摘要
生成模型和推论自动编码器主要在其优化目标中使用$ \ ell_2 $ norm。为了生成感知上更好的图像，这份简短的论文理论上讨论了如何在生成模型和推论自动编码器中使用结构相似性索引（SSIM）。我们首先回顾了SSIM，SSIM距离指标和SSIM内核。我们证明SSIM内核是通用内核，因此可以用于无条件和条件生成的力矩匹配网络中。然后，我们解释了如何在变分和对抗性自动编码器以及无条件和条件生成的对抗网络（GAN）中使用SSIM距离。最后，我们建议在最小二乘gan中使用SSIM距离，而不是$ \ ell_2 $ norm。
Generative models and inferential autoencoders mostly make use of $\ell_2$ norm in their optimization objectives. In order to generate perceptually better images, this short paper theoretically discusses how to use Structural Similarity Index (SSIM) in generative models and inferential autoencoders. We first review SSIM, SSIM distance metrics, and SSIM kernel. We show that the SSIM kernel is a universal kernel and thus can be used in unconditional and conditional generated moment matching networks. Then, we explain how to use SSIM distance in variational and adversarial autoencoders and unconditional and conditional Generative Adversarial Networks (GANs). Finally, we propose to use SSIM distance rather than $\ell_2$ norm in least squares GAN.
