Skip to yearly menu bar Skip to main content


Generating Images with Perceptual Similarity Metrics based on Deep Networks

Alexey Dosovitskiy · Thomas Brox

Area 5+6+7+8 #52

Keywords: [ Deep Learning or Neural Networks ]


We propose a class of loss functions, which we call deep perceptual similarity metrics (DeePSiM), allowing to generate sharp high resolution images from compressed abstract representations. Instead of computing distances in the image space, we compute distances between image features extracted by deep neural networks. This metric reflects perceptual similarity of images much better and, thus, leads to better results. We demonstrate two examples of use cases of the proposed loss: (1) networks that invert the AlexNet convolutional network; (2) a modified version of a variational autoencoder that generates realistic high-resolution random images.

Live content is unavailable. Log in and register to view live content