Image labeling based on graphical models using wasserstein messages and geometric assignment
We introduce a novel approach to Maximum A Posteriori (MAP) inference based on discrete graphical models. By utilizing local Wasserstein distances for coupling assignment measures across edges of the underlying graph, a given discrete objective function is smoothly approximated and restricted to the...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article (Journal) |
| Language: | English |
| Published: |
May 24, 2018
|
| In: |
SIAM journal on imaging sciences
Year: 2018, Volume: 11, Issue: 2, Pages: 1317-1362 |
| ISSN: | 1936-4954 |
| DOI: | 10.1137/17M1150669 |
| Online Access: | Resolving-System, kostenfrei, Volltext: http://dx.doi.org/10.1137/17M1150669 Verlag, kostenfrei, Volltext: https://epubs.siam.org/doi/abs/10.1137/17M1150669 
|
| Author Notes: | Ruben Hühnerbein, Fabrizio Savarino, Freddie Åström, and Christoph Schnörr |
| Summary: | We introduce a novel approach to Maximum A Posteriori (MAP) inference based on discrete graphical models. By utilizing local Wasserstein distances for coupling assignment measures across edges of the underlying graph, a given discrete objective function is smoothly approximated and restricted to the assignment manifold. A corresponding multiplicative update scheme combines in a single process (i) geometric integration of the resulting Riemannian gradient flow, and (ii) rounding to integral solutions that represent valid labelings. Throughout this process, local marginalization constraints known from the established LP relaxation are satisfied, whereas the smooth geometric setting results in rapidly converging iterations that can be carried out in parallel for every edge. |
|---|---|
| Item Description: | Gesehen am 29.08.2018 |
| Physical Description: | Online Resource |
| ISSN: | 1936-4954 |
| DOI: | 10.1137/17M1150669 |