A geometric approach to image labeling
We introduce a smooth non-convex approach in a novel geometric framework which complements established convex and non-convex approaches to image labeling. The major underlying concept is a smooth manifold of probabilistic assignments of a prespecified set of prior data (the “labels”) to given image...
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Dokumenttyp: | Kapitel/Artikel Konferenzschrift |
| Sprache: | Englisch |
| Veröffentlicht: |
16 September 2016
|
| In: |
Computer Vision – ECCV 2016
Year: 2016, Pages: 139-154 |
| DOI: | 10.1007/978-3-319-46454-1_9 |
| Schlagworte: | |
| Online-Zugang: | Verlag, Volltext: http://dx.doi.org/10.1007/978-3-319-46454-1_9 Verlag, Volltext: https://link.springer.com/chapter/10.1007/978-3-319-46454-1_9 
|
| Verfasserangaben: | Freddie Åström, Stefania Petra, Bernhard Schmitzer, Christoph Schnörr |
| Zusammenfassung: | We introduce a smooth non-convex approach in a novel geometric framework which complements established convex and non-convex approaches to image labeling. The major underlying concept is a smooth manifold of probabilistic assignments of a prespecified set of prior data (the “labels”) to given image data. The Riemannian gradient flow with respect to a corresponding objective function evolves on the manifold and terminates, for any δ>0δ>0\delta > 0, within a δδ\delta -neighborhood of an unique assignment (labeling). As a consequence, unlike with convex outer relaxation approaches to (non-submodular) image labeling problems, no post-processing step is needed for the rounding of fractional solutions. Our approach is numerically implemented with sparse, highly-parallel interior-point updates that efficiently converge, largely independent from the number of labels. Experiments with noisy labeling and inpainting problems demonstrate competitive performance. |
|---|---|
| Beschreibung: | Gesehen am 13.03.2018 |
| Beschreibung: | Online Resource |
| ISBN: | 9783319464541 |
| DOI: | 10.1007/978-3-319-46454-1_9 |