NAM: Non-Adversarial Unsupervised Domain Mapping

06/03/2018
by   Yedid Hoshen, et al.
0

Several methods were recently proposed for the task of translating images between domains without prior knowledge in the form of correspondences. The existing methods apply adversarial learning to ensure that the distribution of the mapped source domain is indistinguishable from the target domain, which suffers from known stability issues. In addition, most methods rely heavily on "cycle" relationships between the domains, which enforce a one-to-one mapping. In this work, we introduce an alternative method: Non-Adversarial Mapping (NAM), which separates the task of target domain generative modeling from the cross-domain mapping task. NAM relies on a pre-trained generative model of the target domain, and aligns each source image with an image synthesized from the target domain, while jointly optimizing the domain mapping function. It has several key advantages: higher quality and resolution image translations, simpler and more stable training and reusable target models. Extensive experiments are presented validating the advantages of our method.

READ FULL TEXT

page 8

page 9

page 10

page 12

page 13

research
04/06/2023

Zero-shot Generative Model Adaptation via Image-specific Prompt Learning

Recently, CLIP-guided image synthesis has shown appealing performance on...
research
12/10/2019

Learning Domain Adaptive Features with Unlabeled Domain Bridges

Conventional cross-domain image-to-image translation or unsupervised dom...
research
08/21/2019

TUNA-Net: Task-oriented UNsupervised Adversarial Network for Disease Recognition in Cross-Domain Chest X-rays

In this work, we exploit the unsupervised domain adaptation problem for ...
research
11/15/2022

Dynamic-Pix2Pix: Noise Injected cGAN for Modeling Input and Target Domain Joint Distributions with Limited Training Data

Learning to translate images from a source to a target domain with appli...
research
05/19/2023

Advancing Front Mapping

We present Advancing Front Mapping (AFM), a provably robust algorithm fo...
research
07/23/2018

Generalization Bounds for Unsupervised Cross-Domain Mapping with WGANs

The recent empirical success of cross-domain mapping algorithms, between...
research
03/17/2023

MODIFY: Model-driven Face Stylization without Style Images

Existing face stylization methods always acquire the presence of the tar...

Please sign up or login with your details

Forgot password? Click here to reset