Improving Shape Deformation in Unsupervised Image-to-Image Translation

08/13/2018
by   Aaron Gokaslan, et al.
14

Unsupervised image-to-image translation techniques are able to map local texture between two domains, but they are typically unsuccessful when the domains require larger shape change. Inspired by semantic segmentation, we introduce a discriminator with dilated convolutions that is able to use information from across the entire image to train a more context-aware generator. This is coupled with a multi-scale perceptual loss that is better able to represent error in the underlying shape of objects. We demonstrate that this design is more capable of representing shape deformation in a challenging toy dataset, plus in complex mappings with significant dataset variation between humans, dolls, and anime faces, and between cats and dogs.

READ FULL TEXT

page 2

page 11

page 12

page 13

page 15

page 21

page 22

page 23

research
03/30/2021

SPatchGAN: A Statistical Feature Based Discriminator for Unsupervised Image-to-Image Translation

For unsupervised image-to-image translation, we propose a discriminator ...
research
10/21/2021

A Domain Gap Aware Generative Adversarial Network for Multi-domain Image Translation

Recent image-to-image translation models have shown great success in map...
research
06/06/2018

Unsupervised Attention-guided Image to Image Translation

Current unsupervised image-to-image translation techniques struggle to f...
research
12/30/2021

Leveraging in-domain supervision for unsupervised image-to-image translation tasks via multi-stream generators

Supervision for image-to-image translation (I2I) tasks is hard to come b...
research
05/07/2021

Contrastive Learning for Unsupervised Image-to-Image Translation

Image-to-image translation aims to learn a mapping between different gro...
research
12/19/2017

ComboGAN: Unrestrained Scalability for Image Domain Translation

This year alone has seen unprecedented leaps in the area of learning-bas...
research
11/24/2019

Breaking the cycle – Colleagues are all you need

This paper proposes a novel approach to performing image-to-image transl...

Please sign up or login with your details

Forgot password? Click here to reset