Extremal Domain Translation with Neural Optimal Transport

Milena Gazdieva, Alexander Korotin, Daniil Selikhanovych, Evgeny Burnaev
1/30/2023
cs.LG

Official code: milenagazdieva/ExtremalNeuralOptimalTransport (25 stars, MIT, last commit 2024-12-24).

Abstract

In many unpaired image domain translation problems, e.g., style transfer or super-resolution, it is important to keep the translated image similar to its respective input image. We propose the extremal transport (ET) which is a mathematical formalization of the theoretically best possible unpaired translation between a pair of domains w.r.t. the given similarity function. Inspired by the recent advances in neural optimal transport (OT), we propose a scalable algorithm to approximate ET maps as a limit of partial OT maps. We test our algorithm on toy examples and on the unpaired image-to-image translation task. The code is publicly available at https://github.com/milenagazdieva/ExtremalNeuralOptimalTransport

View on arXivView PDF

Code Implementations(1)

milenagazdieva/ExtremalNeuralOptimalTransportOfficial100%
MIT

PyTorch implementation of "Extremal Domain Translation with Neural Optimal Transport" (NeurIPS 2023)

253Python, Jupyter NotebookNov 7, 20231 years agoMIT
deep-learninggenerative-modelimage-to-image-translationneural-networksoptimal-transport+1 more

Similar Papers

Improving Neural Optimal Transport via Displacement Interpolation

83% match2024

Unbalancedness in Neural Monge Maps Improves Unpaired Domain Translation

83% match1 repo182026

Sparse Domain Transfer via Elastic Net Regularization

81% match2024

Displacement-Sparse Neural Optimal Transport

81% match1 repo22025

Computing high-dimensional optimal transport by flow neural networks

81% match1 repo22023

Monge, Bregman and Occam: Interpretable Optimal Transport in High-Dimensions with Feature-Sparse Maps

80% match2023

Cite this paper

@article{gazdieva2023extremal,
  title  = {Extremal Domain Translation with Neural Optimal Transport},
  author = {Milena Gazdieva and Alexander Korotin and Daniil Selikhanovych and Evgeny Burnaev},
  year   = {2023},
  eprint = {2301.12874},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2301.12874v3},
  journal = {NEURIPS 2023 2023}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2301.12874.svg)](https://codeofpaper.com/paper/2301.12874)

Discussion