OCMLApr 2, 2018

On the Computation of Kantorovich-Wasserstein Distances between 2D-Histograms by Uncapacitated Minimum Cost Flows

arXiv:1804.00445v32 citations
AI Analysis

This work addresses efficiency issues in computing Wasserstein distances for applications in computer vision and machine learning, offering a method that reduces computational complexity from O(n^2) to O(n) for specific norms.

The paper tackles the computational challenge of computing Kantorovich-Wasserstein distances between 2D histograms by approximating the transportation problem with an uncapacitated min cost flow on a reduced network of size O(n), providing optimal solutions for 1-norm and ∞-norm distances and a quantitative error estimate for 2-norm distances, and demonstrates scalability and benefits on benchmark grey scale images.

In this work, we present a method to compute the Kantorovich-Wasserstein distance of order one between a pair of two-dimensional histograms. Recent works in Computer Vision and Machine Learning have shown the benefits of measuring Wasserstein distances of order one between histograms with $n$ bins, by solving a classical transportation problem on very large complete bipartite graphs with $n$ nodes and $n^2$ edges. The main contribution of our work is to approximate the original transportation problem by an uncapacitated min cost flow problem on a reduced flow network of size $O(n)$ that exploits the geometric structure of the cost function. More precisely, when the distance among the bin centers is measured with the 1-norm or the $\infty$-norm, our approach provides an optimal solution. When the distance among bins is measured with the 2-norm: (i) we derive a quantitative estimate on the error between optimal and approximate solution; (ii) given the error, we construct a reduced flow network of size $O(n)$. We numerically show the benefits of our approach by computing Wasserstein distances of order one on a set of grey scale images used as benchmark in the literature. We show how our approach scales with the size of the images with 1-norm, 2-norm and $\infty$-norm ground distances, and we compare it with other two methods which are largely used in the literature.

Code Implementations4 repos
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes