K-ary Regression Forests for Continuous Pose and Direction Estimation

In this work, we propose a novel K-ary splitting method for regression trees and incorporate it into the regression forest framework. Unlike standard binary splitting, where the splitting rule is selected from a predefined set of binary splitting rules via trial-and-error, the proposed K-ary splitting method first finds clusters of 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 into a multiclass classification problem. Consequently, our K-ary splitting method enjoys more freedom in choosing the splitting rules, resulting in more efficient tree structures. In addition to the Euclidean target space, we present a variant which can naturally deal with a circular target space by the proper use of circular statistics. We apply the regression forest employing our K-ary splitting to head pose estimation (Euclidean target space) and car direction estimation (circular target space) and demonstrate that the proposed method significantly outperforms state-of-the-art methods as well as regression forests based on standard binary splitting. The code will be available at our website.