• Corpus ID: 222133384

BalaGAN: Cross-Modal Image Translation Between Imbalanced Domains

  title={BalaGAN: Cross-Modal Image Translation Between Imbalanced Domains},
  author={Or Patashnik and Dov Danon and Hao Zhang and Daniel Cohen-Or},
State-of-the-art image-to-image translation methods tend to struggle in an imbalanced domain setting, where one image domain lacks richness and diversity. We introduce a new unsupervised translation network, BalaGAN, specifically designed to tackle the domain imbalance problem. We leverage the latent modalities of the richer domain to turn the image-to-image translation problem, between two imbalanced domains, into a balanced, multi-class, and conditional translation problem, more resembling… 
2 Citations

Figures and Tables from this paper

Leveraging in-domain supervision for unsupervised image-to-image translation tasks via multi-stream generators
This paper proposes splitting the input data according to semantic masks, explicitly guiding the network to different behavior for the different regions of the image, and proposes training a semantic segmentation network along with the translation task, and to leverage this output as a loss term that improves robustness.
3DSNet: Unsupervised Shape-to-Shape 3D Style Transfer
The first learning-based generative approach for style transfer between 3D objects is proposed, which allows to combine the content and style of a source and target 3D model to generate a novel shape that resembles in style the target while retaining the source content.


Cross-Domain Cascaded Deep Feature Translation
This work leverages VGG, which is a classification network, pre-trained with large-scale semantic supervision, and shows that this method is able to translate between different domains, which exhibit significantly different shapes.
Contrastive Learning for Unpaired Image-to-Image Translation
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.
Unsupervised Image-to-Image Translation Networks
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.
U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation
A novel method for unsupervised image- to-image translation, which incorporates a new attention module and a new learnable normalization function in an end-to-end manner, which can translate both images requiring holistic changes and images requiring large shape changes.
Unsupervised K-modal styled content generation
The emergence of deep generative models has recently enabled the automatic generation of massive amounts of graphical content, both in 2D and in 3D. Generative Adversarial Networks (GANs) and style
Multimodal Unsupervised Image-to-Image Translation
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.
Few-Shot Unsupervised Image-to-Image Translation
This model achieves this few-shot generation capability by coupling an adversarial training scheme with a novel network design, and verifies the effectiveness of the proposed framework through extensive experimental validation and comparisons to several baseline methods on benchmark datasets.
Self-Supervised GANs via Auxiliary Rotation Loss
This work allows the networks to collaborate on the task of representation learning, while being adversarial with respect to the classic GAN game, and takes a step towards bridging the gap between conditional and unconditional GANs.
Self-Attention Generative Adversarial Networks
The proposed SAGAN achieves the state-of-the-art results, boosting the best published Inception score from 36.8 to 52.52 and reducing Frechet Inception distance from 27.62 to 18.65 on the challenging ImageNet dataset.
Large Scale GAN Training for High Fidelity Natural Image Synthesis
It is found that applying orthogonal regularization to the generator renders it amenable to a simple "truncation trick," allowing fine control over the trade-off between sample fidelity and variety by reducing the variance of the Generator's input.