Image-to-image Translation via Hierarchical Style Disentanglement
- URL: http://arxiv.org/abs/2103.01456v1
- Date: Tue, 2 Mar 2021 03:43:18 GMT
- Title: Image-to-image Translation via Hierarchical Style Disentanglement
- Authors: Xinyang Li, Shengchuan Zhang, Jie Hu, Liujuan Cao, Xiaopeng Hong,
Xudong Mao, Feiyue Huang, Yongjian Wu, Rongrong Ji
- Abstract summary: 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.
Both qualitative and quantitative results on the CelebA-HQ dataset verify the ability of the proposed HiSD.
- Score: 115.81148219591387
- License: http://creativecommons.org/licenses/by-nc-sa/4.0/
- Abstract: Recently, image-to-image translation has made significant progress in
achieving both multi-label (\ie, translation conditioned on different labels)
and multi-style (\ie, 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.
Related papers
- DSI2I: Dense Style for Unpaired Image-to-Image Translation [70.93865212275412]
Unpaired exemplar-based image-to-image (UEI2I) translation aims to translate a source image to a target image domain with the style of a target image exemplar.
We propose to represent style as a dense feature map, allowing for a finer-grained transfer to the source image without requiring any external semantic information.
Our results show that the translations produced by our approach are more diverse, preserve the source content better, and are closer to the exemplars when compared to the state-of-the-art methods.
arXiv Detail & Related papers (2022-12-26T18:45:25Z) - Towards Robust and Semantically Organised Latent Representations for
Unsupervised Text Style Transfer [6.467090475885798]
We introduce EPAAEs (versading Perturbed Adrial AutoEncoders) which completes this perturbation model.
We empirically show that this (a) produces a better organised latent space that clusters stylistically similar sentences together.
We also extend the text style transfer tasks to NLI datasets and show that these more complex definitions of style are learned best by EPAAE.
arXiv Detail & Related papers (2022-05-04T20:04:24Z) - Unsupervised Image-to-Image Translation with Generative Prior [103.54337984566877]
Unsupervised image-to-image translation aims to learn the translation between two visual domains without paired data.
We present a novel framework, Generative Prior-guided UN Image-to-image Translation (GP-UNIT), to improve the overall quality and applicability of the translation algorithm.
arXiv Detail & Related papers (2022-04-07T17:59:23Z) - Semi-Supervised Image-to-Image Translation using Latent Space Mapping [37.232496213047845]
We introduce a general framework for semi-supervised image translation.
Our main idea is to learn the translation over the latent feature space instead of the image space.
Thanks to the low dimensional feature space, it is easier to find the desired mapping function.
arXiv Detail & Related papers (2022-03-29T05:14:26Z) - Separating Content and Style for Unsupervised Image-to-Image Translation [20.44733685446886]
Unsupervised image-to-image translation aims to learn the mapping between two visual domains with unpaired samples.
We propose to separate the content code and style code simultaneously in a unified framework.
Based on the correlation between the latent features and the high-level domain-invariant tasks, the proposed framework demonstrates superior performance.
arXiv Detail & Related papers (2021-10-27T12:56:50Z) - Unpaired Image-to-Image Translation via Latent Energy Transport [61.62293304236371]
Image-to-image translation aims to preserve source contents while translating to discriminative target styles between two visual domains.
In this paper, we propose to deploy an energy-based model (EBM) in the latent space of a pretrained autoencoder for this task.
Our model is the first to be applicable to 1024$times$1024-resolution unpaired image translation.
arXiv Detail & Related papers (2020-12-01T17:18:58Z) - Structural-analogy from a Single Image Pair [118.61885732829117]
In this paper, we explore the capabilities of neural networks to understand image structure given only a single pair of images, A and B.
We generate an image that keeps the appearance and style of B, but has a structural arrangement that corresponds to A.
Our method can be used to generate high quality imagery in other conditional generation tasks utilizing images A and B only.
arXiv Detail & Related papers (2020-04-05T14:51:10Z) - Semi-supervised Learning for Few-shot Image-to-Image Translation [89.48165936436183]
We propose a semi-supervised method for few-shot image translation, called SEMIT.
Our method achieves excellent results on four different datasets using as little as 10% of the source labels.
arXiv Detail & Related papers (2020-03-30T22:46:49Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.