Incorporating long-range consistency in CNN-based texture generation

Code

• github.com/guillaumebrg/texture_generation
  OfficialIn papernone★ 13
• github.com/ty625911724/texture_generation_master
  tf★ 0
• github.com/anujdutt9/Artistic-Style-Transfer-using-Keras-Tensorflow
  tf★ 0
• github.com/harunshimanto/Neural-Style-Transfer-of-Artistic-Style
  tf★ 0

Abstract

Gatys et al. (2015) showed that pair-wise products of features in a convolutional network are a very effective representation of image textures. We propose a simple modification to that representation which makes it possible to incorporate long-range structure into image generation, and to render images that satisfy various symmetry constraints. We show how this can greatly improve rendering of regular textures and of images that contain other kinds of symmetric structure. We also present applications to inpainting and season transfer.

Tasks

Reproductions