Boosted Markov Networks for Activity Recognition

08/06/2014
by   Truyen Tran, et al.
0

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.

READ FULL TEXT
research
04/11/2018

Maximum likelihood estimation in hidden Markov models with inhomogeneous noise

We consider parameter estimation in hidden finite state space Markov mod...
research
06/06/2019

SparseSense: Human Activity Recognition from Highly Sparse Sensor Data-streams Using Set-based Neural Networks

Batteryless or so called passive wearables are providing new and innovat...
research
04/01/2016

Leaving Some Stones Unturned: Dynamic Feature Prioritization for Activity Detection in Streaming Video

Current approaches for activity recognition often ignore constraints on ...
research
12/25/2013

Joint segmentation of multivariate time series with hidden process regression for human activity recognition

The problem of human activity recognition is central for understanding a...
research
02/03/2015

Hybrid Orthogonal Projection and Estimation (HOPE): A New Framework to Probe and Learn Neural Networks

In this paper, we propose a novel model for high-dimensional data, calle...
research
10/07/2020

Automated Human Activity Recognition by Colliding Bodies Optimization-based Optimal Feature Selection with Recurrent Neural Network

In smart healthcare, Human Activity Recognition (HAR) is considered to b...
research
04/08/2015

Detecting Falls with X-Factor Hidden Markov Models

Identification of falls while performing normal activities of daily livi...

Please sign up or login with your details

Forgot password? Click here to reset