Few-Shot Unsupervised Image-to-Image Translation

@article{Liu2019FewShotUI,
  title={Few-Shot Unsupervised Image-to-Image Translation},
  author={Ming-Yu Liu and Xun Huang and Arun Mallya and Tero Karras and Timo Aila and Jaakko Lehtinen and Jan Kautz},
  journal={2019 IEEE/CVF International Conference on Computer Vision (ICCV)},
  year={2019},
  pages={10550-10559}
}
  • Ming-Yu Liu, Xun Huang, +4 authors J. Kautz
  • Published 2019
  • Computer Science, Mathematics
  • 2019 IEEE/CVF International Conference on Computer Vision (ICCV)
Unsupervised image-to-image translation methods learn to map images in a given class to an analogous image in a different class, drawing on unstructured (non-registered) datasets of images. [...] Key Method Drawing inspiration from the human capability of picking up the essence of a novel object from a small number of examples and generalizing from there, we seek a few-shot, unsupervised image-to-image translation algorithm that works on previously unseen target classes that are specified, at test time, only by…Expand
Semi-Supervised Learning for Few-Shot Image-to-Image Translation
TLDR
This work proposes applying semi-supervised learning via a noise-tolerant pseudo-labeling procedure, and applies a cycle consistency constraint to further exploit the information from unlabeled images, either from the same dataset or external. Expand
Toward Zero-Shot Unsupervised Image-to-Image Translation
TLDR
This work proposes a zero-shot unsupervised image-to-image translation framework to address this limitation of mode collapse, by associating categories with their side information like attributes, and introduces two strategies for exploiting the space spanned by the semantic attributes. Expand
COCO-FUNIT: Few-Shot Unsupervised Image Translation with a Content Conditioned Style Encoder
TLDR
A new few-shot image translation model, COCO-FUNIT, is proposed, which computes the style embedding of the example images conditioned on the input image and a new module called the constant style bias, which shows effectiveness in addressing the content loss problem. Expand
Unsupervised Image to Image Translation
Unsupervised image-to-image translation methods have received a lot of attention in the last few years. Multiple techniques emerged tackling the initial challenge from different perspectives. SomeExpand
Rethinking the Truly Unsupervised Image-to-Image Translation
TLDR
This paper proposes the truly unsupervised image-to-image translation method (TUNIT) that simultaneously learns to separate image domains via an information-theoretic approach and generate corresponding images using the estimated domain labels. Expand
Contrastive Learning for Unsupervised Image-to-Image Translation
TLDR
An unsupervised image-to-image translation method based on contrastive learning to learn a discriminator that differentiates between distinctive styles and let the discriminator supervise a generator to transfer those styles across images. Expand
Contrastive Learning for Unpaired Image-to-Image Translation
TLDR
The framework enables one-sided translation in the unpaired image-to-image translation setting, while improving quality and reducing training time, and can be extended to the training setting where each "domain" is only a single image. Expand
Leveraging Local Domains for Image-to-Image Translation
TLDR
This paper leverages human knowledge about spatial domain characteristics which it refers to as ’local domains’ and demonstrates its benefit for image-to-image translation and shows that all tested proxy tasks are significantly improved, without ever seeing target domain at training. Expand
Show, Attend, and Translate: Unsupervised Image Translation With Self-Regularization and Attention
TLDR
This work constrain the problem with the assumption that the translated image needs to be perceptually similar to the original image and also appears to be drawn from the new domain, and proposes a simple yet effective image translation model consisting of a single generator trained with a self-regularization term and an adversarial term. Expand
F2GAN: Fusing-and-Filling GAN for Few-shot Image Generation
TLDR
A Fusing-and-Filling Generative Adversarial Network (F2GAN) is proposed to generate realistic and diverse images for a new category with only a few images to ensure the diversity of generated images by a mode seeking loss and an interpolation regression loss. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 65 REFERENCES
Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks
TLDR
This work presents an approach for learning to translate an image from a source domain X to a target domain Y in the absence of paired examples, and introduces a cycle consistency loss to push F(G(X)) ≈ X (and vice versa). Expand
Unsupervised Pixel-Level Domain Adaptation with Generative Adversarial Networks
TLDR
This generative adversarial network (GAN)-based method adapts source-domain images to appear as if drawn from the target domain, and outperforms the state-of-the-art on a number of unsupervised domain adaptation scenarios by large margins. Expand
Unsupervised Image-to-Image Translation Networks
TLDR
This work makes a shared-latent space assumption and proposes an unsupervised image-to-image translation framework based on Coupled GANs that achieves state-of-the-art performance on benchmark datasets. Expand
Image-to-Image Translation with Conditional Adversarial Networks
TLDR
Conditional adversarial networks are investigated as a general-purpose solution to image-to-image translation problems and it is demonstrated that this approach is effective at synthesizing photos from label maps, reconstructing objects from edge maps, and colorizing images, among other tasks. Expand
DualGAN: Unsupervised Dual Learning for Image-to-Image Translation
TLDR
A novel dual-GAN mechanism is developed, which enables image translators to be trained from two sets of unlabeled images from two domains, and can even achieve comparable or slightly better results than conditional GAN trained on fully labeled data. Expand
Multimodal Unsupervised Image-to-Image Translation
TLDR
A Multimodal Unsupervised Image-to-image Translation (MUNIT) framework that assumes that the image representation can be decomposed into a content code that is domain-invariant, and a style code that captures domain-specific properties. Expand
Learning from Simulated and Unsupervised Images through Adversarial Training
TLDR
This work develops a method for S+U learning that uses an adversarial network similar to Generative Adversarial Networks (GANs), but with synthetic images as inputs instead of random vectors, and makes several key modifications to the standard GAN algorithm to preserve annotations, avoid artifacts, and stabilize training. Expand
Toward Multimodal Image-to-Image Translation
TLDR
This work aims to model a distribution of possible outputs in a conditional generative modeling setting that helps prevent a many-to-one mapping from the latent code to the output during training, also known as the problem of mode collapse. Expand
StarGAN: Unified Generative Adversarial Networks for Multi-domain Image-to-Image Translation
TLDR
A unified model architecture of StarGAN allows simultaneous training of multiple datasets with different domains within a single network, which leads to StarGAN's superior quality of translated images compared to existing models as well as the novel capability of flexibly translating an input image to any desired target domain. Expand
Unsupervised Multi-Domain Image Translation with Domain-Specific Encoders/Decoders
TLDR
This work proposes a novel and unified framework named Domain-Bank, which consists of a globally shared auto-encoder and domain-specific encoders/decoders, assuming that there is a universal shared-latent space can be projected and shows the comparable (or even better) image translation results over state-of-the-arts on various challenging unsupervised image translation tasks. Expand
...
1
2
3
4
5
...