One-Shot Unsupervised Cross Domain Translation

By Sagie Benaim, Lior Wolf. 2018

140 citations recorded by the source corpus.

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 this https URL

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Continual Training of Language Models for Few-Shot Learning (2022, methodology): “…space of the few-shot task (Triantaﬁllou et al., 2017; Hu et al., 2018), (ii) data-based methods try to augment additional data to the few-shot set (Benaim and Wolf, 2018; Gao et al., 2020), and (iii) algorithm-based solutions try to improve strategies for searching for the best hypothesis.”
Continual Few-shot Relation Learning via Embedding Space Regularization and Data Augmentation (2022, methodology): “To address this, existing methods have explored three different directions: (i) data-based methods use prior knowledge to augment data to the few-shot set (Santoro et al., 2016; Benaim and Wolf, 2018; Gao et al., 2020); (ii) model-based methods reduce the hypothesis space using prior knowledge (Reze”
LFPT5: A Unified Framework for Lifelong Few-shot Language Learning Based on Prompt Tuning of T5 (2021, methodology): “…of the few-shot task (Triantaﬁllou et al., 2017; Hu et al., 2018), ( ii ) data-based methods that try to augment additional data to the few-shot set (Benaim & Wolf, 2018; Gao et al., 2020b), and ( iii ) algorithm-based solutions that aim to improve strategies for searching for the best hypothesis.”
Generalizing from a Few Examples (2019, background): “These include object recognition [35, 36, 82], font style transfer [7], phrase grounding [162], image retrieval [130], object tracking [14], specific object counting in images [162], scene location recognition [74], gesture recognition [102], part labeling [24], image generation [34, 76, 107, 109], ”
Few-Shot Unsupervised Image-to-Image Translation (2019, background): “Recently, Benaim and Wolf [4] proposed an unsupervised image-to-image translation framework for partially addressing the ﬁrst aspect.”
Image-to-Image Translation: Methods and Applications (2021, methodology): “style transfer artistic:[113], [1] artistic:[15] artistic:[2], [36], [136], [137], [145]; photo-realistic: [102], [139], [2], [75], [136], [137], artistic:[152]; photo-realistic: [90], [76], [77], [77], [143], [152], [154] \ artistic:[155], [81], [166]; photo-realistic: [155], [156], [81] artistic:[”
Few-shot Semantic Image Synthesis Using StyleGAN Prior (2021, unknown): “Image-to-Image translation There are various image-to-image (I2I) translation methods suitable for semantic image synthesis; the goals are, e.g ., to improve image qual-Table [14, 23, 22] paired large large none “Few-shot” I2I [21, 36] unpaired large large small Benaim and Wolf [4] unpaired small la”
Few-shot Video-to-Video Synthesis (2019, background): “Image-to-image synthesis , which transfers an input image from one domain to a corresponding image in another domain [22, 50, 3, 46, 68, 30, 21, 69, 58, 8, 41, 31, 2], is the foundation of vid2vid .”