One-Shot Unsupervised Cross Domain Translation

Abstract

Given a single image x from domain A and a set of images from domain B, our task is to generate the analogous of x in B. We argue that this task could be a key AI capability that underlines the ability of cognitive agents to act in the world and present empirical evidence that the existing unsupervised domain translation methods fail on this task. Our method follows a two step process. First, a variational autoencoder for domain B is trained. Then, given the new sample x, we create a variational autoencoder for domain A by adapting the layers that are close to the image in order to directly fit x, and only indirectly adapt the other layers. Our experiments indicate that the new method does as well, when trained on one sample x, as the existing domain transfer methods, when these enjoy a multitude of training samples from domain A. Our code is made publicly available at https://github.com/sagiebenaim/OneShotTranslation

Published at NIPS 2018

References in corpus (7)

• How transferable are features in deep neural networks?
• Unsupervised Image-to-Image Translation Networks
• Unsupervised Cross-Domain Image Generation
• DualGAN: Unsupervised Dual Learning for Image-to-Image Translation
• Word Translation Without Parallel Data
• One-Sided Unsupervised Domain Mapping
• Non-Adversarial Unsupervised Word Translation

Cited by in corpus (28)

• Generalizing from a Few Examples: A Survey on Few-Shot Learning
• Few-Shot Unsupervised Image-to-Image Translation
• Adversarial Style Mining for One-Shot Unsupervised Domain Adaptation
• Generative Adversarial Networks for Image and Video Synthesis: Algorithms and Applications
• Image-to-Image Translation: Methods and Applications
• Geometry-Consistent Generative Adversarial Networks for One-Sided Unsupervised Domain Mapping
• MT3: Meta Test-Time Training for Self-Supervised Test-Time Adaption
• Anomaly Detection with Domain Adaptation
• ZstGAN: An Adversarial Approach for Unsupervised Zero-Shot Image-to-Image Translation
• COCO-FUNIT: Few-Shot Unsupervised Image Translation with a Content Conditioned Style Encoder
• Domain-Scalable Unpaired Image Translation via Latent Space Anchoring
• Few-shot Semantic Image Synthesis Using StyleGAN Prior
• Data Efficient Unsupervised Domain Adaptation for Cross-Modality Image Segmentation
• Zero-Shot Learning with Knowledge Enhanced Visual Semantic Embeddings
• TuiGAN: Learning Versatile Image-to-Image Translation with Two Unpaired Images
• Learning to Transfer: Unsupervised Meta Domain Translation
• One-Shot Unsupervised Cross-Domain Detection
• A Novel BiLevel Paradigm for Image-to-Image Translation
• Self domain adapted network
• Semi-supervised Learning for Few-shot Image-to-Image Translation
• What Can Knowledge Bring to Machine Learning? -- A Survey of Low-shot Learning for Structured Data
• A Quantitative Perspective on Values of Domain Knowledge for Machine Learning
• World-Consistent Video-to-Video Synthesis
• One-Shot Image-to-Image Translation via Part-Global Learning with a Multi-adversarial Framework
• One-shot Unsupervised Domain Adaptation with Personalized Diffusion Models
• Few-shot Knowledge Transfer for Fine-grained Cartoon Face Generation
• Few Shot Learning With No Labels
• Toward Zero-Shot Unsupervised Image-to-Image Translation