Solve optimal transport problems. Compute Wasserstein distances (a.k.a. Kantorovitch, Fortet--Mourier, Mallows, Earth Mover's, or minimal \(L_p\) distances), return the corresponding transport plans, and display them graphically. Objects that can be compared include grey-scale images, (weighted) point patterns, and mass vectors.
Dominic Schuhmacher [email protected]
Björn Bähre [email protected] (code for aha-method)
Nicolas Bonneel [email protected]
(adaptation of LEMON code for fast networkflow method)
Carsten Gottschlich [email protected]
(original java code for shortlist and revsimplex methods)
Valentin Hartmann [email protected] (code for aha method for p=1)
Florian Heinemann [email protected]
(integration of networkflow method)
Bernhard Schmitzer [email protected] (shielding method)
Jörn Schrieber [email protected] (subsampling method)
Maintainer: Dominic Schuhmacher [email protected]
| Package: | transport |
| Type: | Package |
| Version: | 0.12-1 |
| Date: | 2019-08-07 |
| License: | GPL (>=2) |
| LazyData: | yes |
The main end-user function is transport. It computes optimal transport plans between images (class pgrid), point patterns (class pp), weighted point patterns (class wpp) and mass vectors, based on various algorithms. These transport plans can be ploted. The function wasserstein allows for the numerical computation of \(p\)-th order Wasserstein distances.
Most functions in this package are designed for data in two and higher dimensions. A quick tool for computing the \(p\)-th order Wasserstein distance between univariate samples is wasserstein1d.
See help page for the function transport.
## See examples for function transport
Run the code above in your browser using DataLab