Paper Info
Title
Learning the right model: Efficient max-margin learning in Laplacian CRFs
Abstract
An important modeling decision made while designing Conditional Random Fields (CRFs) is the choice of the potential functions over the cliques of variables. Laplacian potentials are useful because they are robust potentials and match image statistics better than Gaussians. Moreover, energies with Laplacian terms remain convex, which simplifies inference. This makes Laplacian potentials an ideal modeling choice for some applications. In this paper, we study max-margin parameter learning in CRFs with Laplacian potentials (LCRFs). We first show that structured hinge-loss [35] is non-convex for LCRFs and thus techniques used by previous works are not applicable. We then present the first approximate max-margin algorithm for LCRFs. Finally, we make our learning algorithm scalable in the number of training images by using dual-decomposition techniques. Our experiments on single-image depth estimation show that even with simple features, our approach achieves comparable to state-of-art results.
Year
Venue
Keywords
2012
CVPR
Laplacian potential,max-margin parameter,important modeling decision,approximate max-margin algorithm,dual-decomposition technique,single-image depth estimation show,ideal modeling choice,algorithm scalable,right model,Laplacian term,efficient max-margin,Laplacian CRFs,Conditional Random Fields
DocType
Citations 
PageRank 
Conference
2
0.57
References 
Authors
0
1
Name
Order
Citations
PageRank
Ashutosh Saxena14575227.88