Instance Normalization: The Missing Ingredient for Fast Stylization

TLDR

The fast stylization technique was originally introduced by Ulyanov et al. (2016). This paper revisits that fast stylization method to explore potential improvements. The authors replace batch normalization with instance normalization during both training and testing. The substitution yields a noticeable qualitative improvement in generated images and enables training of high‑performance real‑time image generation architectures.

Abstract

It this paper we revisit the fast stylization method introduced in Ulyanov et. al. (2016). We show how a small change in the stylization architecture results in a significant qualitative improvement in the generated images. The change is limited to swapping batch normalization with instance normalization, and to apply the latter both at training and testing times. The resulting method can be used to train high-performance architectures for real-time image generation. The code will is made available on github at https://github.com/DmitryUlyanov/texture_nets. Full paper can be found at arXiv:1701.02096.