Image-to-image Translation via Hierarchical Style Disentanglement

03/02/2021 ∙ by Xinyang Li, et al. ∙ 5

Recently, image-to-image translation has made significant progress in achieving both multi-label (, translation conditioned on different labels) and multi-style (, generation with diverse styles) tasks. However, due to the unexplored independence and exclusiveness in the labels, existing endeavors are defeated by involving uncontrolled manipulations to the translation results. In this paper, we propose Hierarchical Style Disentanglement (HiSD) to address this issue. Specifically, we organize the labels into a hierarchical tree structure, in which independent tags, exclusive attributes, and disentangled styles are allocated from top to bottom. Correspondingly, a new translation process is designed to adapt the above structure, in which the styles are identified for controllable translations. Both qualitative and quantitative results on the CelebA-HQ dataset verify the ability of the proposed HiSD. We hope our method will serve as a solid baseline and provide fresh insights with the hierarchically organized annotations for future research in image-to-image translation. The code has been released at https://github.com/imlixinyang/HiSD.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 2

page 6

page 11

page 12

page 13

page 14

page 15

page 16

Code Repositories

HiSD

Official pytorch implementation of paper "Image-to-image Translation via Hierarchical Style Disentanglement" (CVPR 2021 Oral).


view repo
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.