Meta Internal Learning

by   Raphael Bensadoun, et al.

Internal learning for single-image generation is a framework, where a generator is trained to produce novel images based on a single image. Since these models are trained on a single image, they are limited in their scale and application. To overcome these issues, we propose a meta-learning approach that enables training over a collection of images, in order to model the internal statistics of the sample image more effectively. In the presented meta-learning approach, a single-image GAN model is generated given an input image, via a convolutional feedforward hypernetwork f. This network is trained over a dataset of images, allowing for feature sharing among different models, and for interpolation in the space of generative models. The generated single-image model contains a hierarchy of multiple generators and discriminators. It is therefore required to train the meta-learner in an adversarial manner, which requires careful design choices that we justify by a theoretical analysis. Our results show that the models obtained are as suitable as single-image GANs for many common image applications, significantly reduce the training time per image without loss in performance, and introduce novel capabilities, such as interpolation and feedforward modeling of novel images.


page 21

page 23

page 25

page 26

page 28

page 35

page 36

page 37


Training End-to-end Single Image Generators without GANs

We present AugurOne, a novel approach for training single image generati...

Improved Techniques for Training Single-Image GANs

Recently there has been an interest in the potential of learning generat...

Unsupervised 3D Reconstruction from a Single Image via Adversarial Learning

Recent advancements in deep learning opened new opportunities for learni...

180-degree Outpainting from a Single Image

Presenting context images to a viewer's peripheral vision is one of the ...

MOGAN: Morphologic-structure-aware Generative Learning from a Single Image

In most interactive image generation tasks, given regions of interest (R...

PetsGAN: Rethinking Priors for Single Image Generation

Single image generation (SIG), described as generating diverse samples t...

BlendGAN: Learning and Blending the Internal Distributions of Single Images by Spatial Image-Identity Conditioning

Training a generative model on a single image has drawn significant atte...

Please sign up or login with your details

Forgot password? Click here to reset