Perceptual Losses for Real-Time Style Transfer and Super-Resolution

Justin Johnson, Alexandre Alahi, Li Fei-Fei

Abstract

We consider image transformation problems, where an input image is transformed into an output image. Recent methods for such problems typically train feed-forward convolutional neural networks using a per-pixel loss between the output and ground-truth images. Parallel work has shown that high-quality images can be generated by defining and optimizing perceptual loss functions based on high-level features extracted from pretrained networks. We combine the benefits of both approaches, and propose the use of perceptual loss functions for training feed-forward networks for image transformation tasks. We show results on image style transfer, where a feed-forward network is trained to solve the optimization problem proposed by Gatys et al. in real-time. Compared to the optimization-based method, our network gives similar qualitative results but is three orders of magnitude faster. We also experiment with single-image super-resolution, where replacing a per-pixel loss with a perceptual loss gives visually pleasing results.
BibTeX
@inproceedings{johnson2016perceptual,
  author = {Justin Johnson and Alexandre Alahi and Li Fei-Fei},
  title = {{Perceptual Losses for Real-Time Style Transfer and Super-Resolution}},
  booktitle = {European Conference on Computer Vision},
  year = {2016},
  doi = {10.1007/978-3-319-46475-6\_43},
}
Figure