Boosted Markov Networks for Activity Recognition

TL;DR

This paper introduces boosted Markov networks with hidden variables for activity recognition, combining boosting with Markov networks to improve model sparsity and computational efficiency.

Contribution

It extends boosted Markov networks to include hidden variables and demonstrates their effectiveness in activity recognition and feature selection.

Findings

Comparable performance to maximum likelihood estimation

Learns sparse models for computational savings

Effective in video-based activity recognition

Abstract

We explore a framework called boosted Markov networks to combine the learning capacity of boosting and the rich modeling semantics of Markov networks and applying the framework for video-based activity recognition. Importantly, we extend the framework to incorporate hidden variables. We show how the framework can be applied for both model learning and feature selection. We demonstrate that boosted Markov networks with hidden variables perform comparably with the standard maximum likelihood estimation. However, our framework is able to learn sparse models, and therefore can provide computational savings when the learned models are used for classification.