FIGR: Few-shot Image Generation with Reptile

By Louis Clouâtre, Marc Demers. 2019

111 citations recorded by the source corpus.

Generative Adversarial Networks (GAN) boast impressive capacity to generate realistic images. However, like much of the field of deep learning, they require an inordinate amount of data to produce results, thereby limiting their usefulness in generating novelty. In the same vein, recent advances in meta-learning have opened the door to many few-shot learning applications. In the present work, we propose Few-shot Image Generation using Reptile (FIGR), a GAN meta-trained with Reptile. Our model successfully generates novel images on both MNIST and Omniglot with as little as 4 images from an unseen class. We further contribute FIGR-8, a new dataset for few-shot image generation, which contains 1,548,944 icons categorized in over 18,409 classes. Trained on FIGR-8, initial results show that our model can generalize to more advanced concepts (such as "bird" and "knife") from as few as 8 samples from a previously unseen class of images and as little as 10 training steps through those 8 images. This work demonstrates the potential of training a GAN for few-shot image generation and aims to set a new benchmark for future work in the domain.

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

Meta-StyleSpeech : Multi-Speaker Adaptive Text-to-Speech Generation (2021, methodology): “In this regard, some of the few-shot generative models have utilized meta-learning for improved generalization (Rezende et al., 2016; Bartunov & Vetrov, 2018; Clouˆatre & Demers, 2019).”
Watch Your Up-Convolution: CNN Based Generative Deep Neural Networks Are Failing to Reproduce Spectral Distributions (2020, background): “…images [29, 6], image-to-image [45, 26, 61, 9, 42, 30] and text-to-image translations [48, 11, 58, 59], style transfer [27, 60, 61, 25], image inpainting [45, 54, 33, 26, 56], transfer learning [5, 10, 15] or even for training semantic segmentation tasks [35, 53], just to name a few.”
A Meta-Learning Framework for Generalized Zero-Shot Learning (2019, methodology): “We would also like to highlight that, although we develop this model with the focus being ZSL and generalized ZSL, our ideas can be used for the task of supervised few-shot generation [8], which is the problem of learning to generate data given very few examples to learn the data distribution.”