Continuous Methods : Hamiltonian Domain Translation
Computer Vision and Pattern Recognition
2022-07-11 v1
Abstract
This paper proposes a novel approach to domain translation. Leveraging established parallels between generative models and dynamical systems, we propose a reformulation of the Cycle-GAN architecture. By embedding our model with a Hamiltonian structure, we obtain a continuous, expressive and most importantly invertible generative model for domain translation.
Cite
@article{arxiv.2207.03843,
title = {Continuous Methods : Hamiltonian Domain Translation},
author = {Emmanuel Menier and Michele Alessandro Bucci and Mouadh Yagoubi and Lionel Mathelin and Marc Schoenauer},
journal= {arXiv preprint arXiv:2207.03843},
year = {2022}
}