Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

Learning to Generate Wasserstein Barycenters

Julien Lacombe Julie Digne Nicolas Courty Nicolas Bonneel 1
1 Origami - Origami
LIRIS - Laboratoire d'InfoRmatique en Image et Systèmes d'information
Abstract : Optimal transport is a notoriously difficult problem to solve numerically, with current approaches often remaining intractable for very large scale applications such as those encountered in machine learning. Wasserstein barycenters -- the problem of finding measures in-between given input measures in the optimal transport sense -- is even more computationally demanding as it requires to solve an optimization problem involving optimal transport distances. By training a deep convolutional neural network, we improve by a factor of 60 the computational speed of Wasserstein barycenters over the fastest state-of-the-art approach on the GPU, resulting in milliseconds computational times on $512\times512$ regular grids. We show that our network, trained on Wasserstein barycenters of pairs of measures, generalizes well to the problem of finding Wasserstein barycenters of more than two measures. We demonstrate the efficiency of our approach for computing barycenters of sketches and transferring colors between multiple images.
Document type :
Preprints, Working Papers, ...
Complete list of metadata
Contributor : Nicolas Bonneel Connect in order to contact the contributor
Submitted on : Tuesday, October 12, 2021 - 10:29:20 AM
Last modification on : Friday, October 15, 2021 - 3:43:46 AM


Files produced by the author(s)


  • HAL Id : hal-03374452, version 1
  • ARXIV : 2102.12178


Julien Lacombe, Julie Digne, Nicolas Courty, Nicolas Bonneel. Learning to Generate Wasserstein Barycenters. 2021. ⟨hal-03374452⟩



Record views


Files downloads