TY - GEN
T1 - ADABOOST GPU-based classifier for direct volume rendering
AU - Amoros, Oscar
AU - Escalera, Sergio
AU - Puig, Anna
PY - 2011
Y1 - 2011
N2 - In volume visualization, the voxel visibility and materials are carried out through an interactive editing of Transfer Function. In this paper, we present a two-level GPU-based labeling method that computes in times of rendering a set of labeled structures using the Adaboost machine learning classifier. In a pre-processing step, Adaboost trains a binary classifier from a pre-labeled dataset and, in each sample, takes into account a set of features. This binary classifier is a weighted combination of weak classifiers, which can be expressed as simple decision functions estimated on a single feature values. Then, at the testing stage, each weak classifier is independently applied on the features of a set of unlabeled samples. We propose an alternative representation of these classifiers that allow a GPU-based parallelizated testing stage embedded into the visualization pipeline. The empirical results confirm the OpenCL-based classification of biomedical datasets as a tough problem where an opportunity for further research emerges.
AB - In volume visualization, the voxel visibility and materials are carried out through an interactive editing of Transfer Function. In this paper, we present a two-level GPU-based labeling method that computes in times of rendering a set of labeled structures using the Adaboost machine learning classifier. In a pre-processing step, Adaboost trains a binary classifier from a pre-labeled dataset and, in each sample, takes into account a set of features. This binary classifier is a weighted combination of weak classifiers, which can be expressed as simple decision functions estimated on a single feature values. Then, at the testing stage, each weak classifier is independently applied on the features of a set of unlabeled samples. We propose an alternative representation of these classifiers that allow a GPU-based parallelizated testing stage embedded into the visualization pipeline. The empirical results confirm the OpenCL-based classification of biomedical datasets as a tough problem where an opportunity for further research emerges.
KW - Computing and parallel rendering
KW - High-performance
KW - Rendering hardware
KW - Volume rendering
UR - http://www.scopus.com/inward/record.url?scp=79960228124&partnerID=8YFLogxK
M3 - Article in proceeding
AN - SCOPUS:79960228124
SN - 9789898425454
T3 - GRAPP 2011 - Proceedings of the International Conference on Computer Graphics Theory and Applications
SP - 215
EP - 219
BT - GRAPP 2011 - Proceedings of the International Conference on Computer Graphics Theory and Applications
T2 - International Conference on Computer Graphics Theory and Applications, GRAPP 2011
Y2 - 5 March 2011 through 7 March 2011
ER -