Dynamic hybrid algorithms for MAP inference in discrete MRFs
<p>In this paper, we present novel techniques that improve the computational and memory efficiency of algorithms for solving multi-label energy functions arising from discrete MRFs or CRFs. These methods are motivated by the observations that the performance of minimization algorithms depends...
Main Authors: | Alahari, K, Kohli, P, Torr, PHS |
---|---|
Formato: | Journal article |
Idioma: | English |
Publicado em: |
IEEE
2009
|
Registos relacionados
-
Reduce, reuse and recycle: efficiently solving multi-label MRFs
Por: Alahari, K, et al.
Publicado em: (2008) -
An analysis of convex relaxations for MAP estimation of discrete MRFs
Por: Pawan Kumar, M, et al.
Publicado em: (2009) -
A tiered move-making algorithm for general pairwise MRFs
Por: Vineet, V, et al.
Publicado em: (2012) -
On partial optimality in multi-label MRFs
Por: Kohli, P, et al.
Publicado em: (2008) -
Exact inference in multi-label CRFs with higher order cliques
Por: Ramalingam, S, et al.
Publicado em: (2008)