A pragmatic approach to multi-class classification

TL;DR

This paper introduces a flexible hierarchical multi-class classification method that enhances existing models by leveraging classifier outputs, demonstrated on a complex gesture recognition task.

Contribution

It proposes a generic cascade approach that improves multi-class classification without relying on probabilistic assumptions, applicable to various models.

Findings

Effective on a ten-class 3D gesture recognition task

Harnesses correlations between class predictions

Applicable to multiple classification models

Abstract

We present a novel hierarchical approach to multi-class classification which is generic in that it can be applied to different classification models (e.g., support vector machines, perceptrons), and makes no explicit assumptions about the probabilistic structure of the problem as it is usually done in multi-class classification. By adding a cascade of additional classifiers, each of which receives the previous classifier's output in addition to regular input data, the approach harnesses unused information that manifests itself in the form of, e.g., correlations between predicted classes. Using multilayer perceptrons as a classification model, we demonstrate the validity of this approach by testing it on a complex ten-class 3D gesture recognition task.