TY  - JOUR
AU  - Francesco Ciompi
AU  - Oriol Pujol
AU  - Petia Radeva
PY  - 2014//
TI  - ECOC-DRF: Discriminative random fields based on error correcting output codes
T2  - PR
JO  - Pattern Recognition
SP  - 2193
EP  - 2204
VL  - 47
IS  - 6
KW  - Discriminative random fields
KW  - Error-correcting output codes
KW  - Multi-class classification
KW  - Graphical models
N2  - We present ECOC-DRF, a framework where potential functions for Discriminative Random Fields are formulated as an ensemble of classifiers. We introduce the label trick, a technique to express transitions in the pairwise potential as meta-classes. This allows to independently learn any possible transition between labels without assuming any pre-defined model. The Error Correcting Output Codes matrix is used as ensemble framework for the combination of margin classifiers. We apply ECOC-DRF to a large set of classification problems, covering synthetic, natural and medical images for binary and multi-class cases, outperforming state-of-the art in almost all the experiments.
UR  - http://dx.doi.org/10.1016/j.patcog.2013.12.007
N1  - LAMP; HuPBA; MILAB; 605.203; 600.046; 601.043; 600.079
ID  - Francesco Ciompi2014
ER  -