Few-Shot Unsupervised Image-to-Image Translation
This addresses a limitation in image-to-image translation for applications needing adaptation to new classes with minimal data, though it is incremental as it builds on existing adversarial training methods.
The paper tackles the problem of unsupervised image-to-image translation requiring many images per class by proposing a few-shot method that works on unseen target classes with only a few examples, achieving effective results as verified through experiments on benchmark datasets.
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. While remarkably successful, current methods require access to many images in both source and destination classes at training time. We argue this greatly limits their use. 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 a few example images. Our model achieves this few-shot generation capability by coupling an adversarial training scheme with a novel network design. Through extensive experimental validation and comparisons to several baseline methods on benchmark datasets, we verify the effectiveness of the proposed framework. Our implementation and datasets are available at https://github.com/NVlabs/FUNIT .