AdaIN: a simple and very powerful idea for neural style transfer

Released in: Arbitrary Style Transfer in Real-Time With Adaptive Instance Normalization



Gatys et al. recently introduced a neural algorithm that renders a content image in the style of another image, achieving so-called style transfer. However, their framework requires a slow iterative optimization process, which limits its practical application. Fast approximations with feed-forward neural networks have been proposed to speed up neural style transfer. Unfortunately, the speed improvement comes at a cost: the network is usually tied to a fixed set of styles and cannot adapt to arbitrary new styles. In this paper, the authors present a simple yet effective approach that for the first time enables arbitrary style transfer in real-time. At the heart of the proposed method is a novel adaptive instance normalization (AdaIN) layer that aligns the mean and variance of the content features with those of the style features. AdaIN achieves speed comparable to the fastest existing approach, without the restriction to a pre-defined set of styles. In addition, the proposed approach allows flexible user controls such as content-style trade-off, style interpolation, color & spatial controls, all using a single feed-forward neural network.


Year Released

Key Links & Stats

Style Transfer Module

Arbitrary Style Transfer in Real-Time With Adaptive Instance Normalization

@InProceedings{Huang_2017_ICCV, author = {Huang, Xun and Belongie, Serge}, title = {Arbitrary Style Transfer in Real-Time With Adaptive Instance Normalization}, booktitle = {Proceedings of the IEEE International Conference on Computer Vision (ICCV)}, month = {Oct}, year = {2017} }

ML Tasks

  1. General
  2. Domain Adaptation
  3. Style Transfer

ML Platform

  1. Pytorch
  2. Tensorflow


  1. General
  2. Still Image


  1. General

CG Platform

  1. Not Applicable

Related organizations

Cornell University