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...
Auteurs principaux: | Kohli, P, Ladický, L, Torr, PHS |
---|---|
Format: | Journal article |
Langue: | English |
Publié: |
Springer
2009
|
Documents similaires
-
Robust higher order potentials for enforcing label consistency
par: Kohli, P, et autres
Publié: (2008) -
Exact inference in multi-label CRFs with higher order cliques
par: Ramalingam, S, et autres
Publié: (2008) -
P3 and beyond: solving energies with higher order cliques
par: Kohli, P, et autres
Publié: (2007) -
Associative hierarchical CRFs for object class image segmentation
par: Ladický, L, et autres
Publié: (2009) -
Exact and approximate inference in associative hierarchical networks using graph cuts
par: Russell, C, et autres
Publié: (2010)