Boosting and structure learning in dynamic Bayesian networks for audio-visual speaker detection
Published 25 June 2003
Tanzeem Choudhury, James M. Rehg, Vladimir Pavlović, Alex Pentland
Citations47
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
This paper introduces a new boosted structure learning algorithm based on AdaBoost that modifies both the network structure and parameters so as to improve classification accuracy and presents results for speaker detection and for the UCI "chess" dataset.
Abstract
Bayesian networks are an attractive modeling tool for human sensing, as they combine an intuitive graphical representation with efficient algorithms for inference and learning. Earlier work has demonstrated that boosted parameter learning could be used to improve the performance of Bayesian network classifiers for complex multi-modal inference problems such as speaker detection. In speaker detection, the goal is to use video and audio cues to infer when a person is speaking to a user interface. In this paper we introduce a new boosted structure learning algorithm based on AdaBoost. Given labeled data, our algorithm modifies both the network structure and parameters so as to improve classification accuracy. We compare its performance to both standard structure learning and boosted parameter learning on a fixed structure. We present results for speaker detection and for the UCI "chess" dataset.
Keywords
Computer Science
Machine LearningBayesian Network Classifiers
4,738 Citations1997Nir Friedman, Dan Geiger +1 more
Tree Augmented Naive Bayes (TAN) is single out, which outperforms naive Bayes, yet at the same time maintains the computational simplicity and robustness that characterize naive Baye.
Machine LearningA Bayesian Method for the Induction of Probabilistic Networks from Data
3,525 Citations1992Gregory F. Cooper, Edward H. Herskovits
This paper presents a Bayesian method for constructing probabilistic networks from databases, focusing on constructing Bayesian belief networks, and extends the basic method to handle missing data and hidden variables.
IEEE Transactions on Pattern Analysis and Machine IntelligenceNeural network-based face detection
3,521 Citations1998Henry A. Rowley, Shumeet Baluja +1 more
Introduction to Bayesian Networks
2,585 Citations2019
Improved boosting algorithms using confidence-rated predictions
2,568 Citations1998Robert E. Schapire, Yoram Singer
Machine LearningA Bayesian method for the induction of probabilistic networks from data
2,344 Citations1992Gregory F. Cooper, Edward H. Herskovits
This paper presents a Bayesian method for constructing probabilistic networks from databases, focusing on constructing Bayesian belief networks, and extends the basic method to handle missing data and hidden variables.
Machine LearningImproved Boosting Algorithms Using Confidence-rated Predictions
1,959 Citations1999Robert E. Schapire, Yoram Singer
Several improvements to Freund and Schapire's AdaBoost boosting algorithm are described, particularly in a setting in which hypotheses may assign confidences to each of their predictions.
Studies in computational intelligenceA Tutorial on Learning with Bayesian Networks
1,699 Citations2008David Heckerman
Methods for constructing Bayesian networks from prior knowledge are discussed and methods for using data to improve these models are summarized, including techniques for learning with incomplete data.
Coupled hidden Markov models for complex action recognition
1,059 Citations2002Matthew Brand, Nuria Oliver +1 more
Algorithms for coupling and training hidden Markov models (HMMs) to model interacting processes, and demonstrate their superiority to conventional HMMs in a vision task classifying two-handed actions are presented.
arXiv (Cornell University)Learning the Structure of Dynamic Probabilistic Networks
574 Citations2013Nir Friedman, Kevin J. Murphy +1 more
arXiv (Cornell University)Object-Oriented Bayesian Networks
497 Citations2013Daphne Koller, Avi Pfeffer
arXiv (Cornell University)Being Bayesian about Network Structure
267 Citations2013Nir Friedman, Daphne Koller
This paper shows how to efficiently compute a sum over the exponential number of networks that are consistent with a fixed ordering over network variables, and uses this result as the basis for an algorithm that approximates the Bayesian posterior of a feature.
Learning Joint Statistical Models for Audio-Visual Fusion and Segregation
182 Citations2000John W. Fisher, Trevor Darrell +2 more
First, the data is projected into a maximally informative, low-dimensional subspace, suitable for density estimation, and the complicated stochastic relationships between the signals are modeled using a nonparametric density estimator.
Look who's talking: speaker detection using video and audio correlation
137 Citations2002Ross Cutler, L.S. Davis
A method of automatically detecting a talking person using video and audio data from a single microphone using a time-delayed neural network and a spatio-temporal search for a speaking person is described.
Vision for a smart kiosk
92 Citations2002James M. Rehg, Michael Loughlin +1 more
A novel computer vision application: vision-based human sensing for a Smart Kiosk interface and a prototype kiosk which uses color stereo tracking and graphical output to interact with several users is described.
Vision-based speaker detection using Bayesian networks
65 Citations2003James M. Rehg, Kevin P. Murphy +1 more
It is argued that Bayesian network models are an attractive statistical framework for cue fusion in these applications because they combine a natural mechanism for expressing contextual information with efficient algorithms for learning and inference.
Multimodal speaker detection using error feedback dynamic Bayesian networks
53 Citations2002Vladimir Pavlović, A. Garg +2 more
This work forms a learning framework for DBNs based on error-feedback and statistical boosting theory and applies this framework to the problem of audio/visual speaker detection in an interactive kiosk environment using "off-the-shelf" visual and audio sensors.
Audio-visual speaker detection using dynamic Bayesian networks
42 Citations2002A. Garg, Vladimir Pavlović +1 more
This work demonstrates the use of dynamic Bayesian networks (DBN) in tackling the problem of audio/visual speaker detection by optimally fused visual and audio sensors along with contextual information in a DBN architecture that infers instances when an individual is speaking.