Robust higher order potentials for enforcing label consistency
This paper proposes a novel framework for labelling problems which is able to combine multiple segmentations in a principled manner. Our method is based on higher order conditional random fields and uses potentials defined on sets of pixels (image segments) generated using unsupervised segmentation...
Autori principali: | Kohli, P, Ladický, L, Torr, PHS |
---|---|
Natura: | Journal article |
Lingua: | English |
Pubblicazione: |
Springer
2009
|
Documenti analoghi
-
Robust higher order potentials for enforcing label consistency
di: Kohli, P, et al.
Pubblicazione: (2008) -
Exact inference in multi-label CRFs with higher order cliques
di: Ramalingam, S, et al.
Pubblicazione: (2008) -
P3 and beyond: solving energies with higher order cliques
di: Kohli, P, et al.
Pubblicazione: (2007) -
Associative hierarchical CRFs for object class image segmentation
di: Ladický, L, et al.
Pubblicazione: (2009) -
Exact and approximate inference in associative hierarchical networks using graph cuts
di: Russell, C, et al.
Pubblicazione: (2010)