Growing Regression Tree Forests by Classification

Authors: Kota Hara Rama Chellappa

Publish Date: 2016/08/27

Volume: 122, Issue: 2, Pages: 292-312

PDF Link

Abstract

We propose a novel node splitting method for regression trees and incorporate it into the random regression forest framework Unlike traditional binary splitting where the splitting rule is selected from a predefined set of binary splitting rules via trialanderror the proposed node splitting method first finds clusters in the training data which at least locally minimize the empirical loss without considering the input space Then splitting rules which preserve the found clusters as much as possible are determined by casting the problem as a classification problem Consequently our new node splitting method enjoys more freedom in choosing the splitting rules resulting in more efficient tree structures In addition to the algorithm for the ordinary Euclidean target space we present a variant which can naturally deal with a circular target space by the proper use of circular statistics In order to deal with challenging ambiguous imagebased pose estimation problems we also present a votingbased ensemble method using the mean shift algorithm Furthermore to address data imbalanceness problems present in some of the datasets we propose a bootstrap sampling method using a sample weighting technique We apply the proposed random regression forest algorithm to head pose estimation car direction estimation and pedestrian orientation estimation tasks and demonstrate its competitive performance

Keywords: