Decomposition in hidden Markov models for activity recognition

Authors:
Weidong Zhang;Feng Chen;Wenli Xu;Zisheng Cao
Affiliations:
Department of Automation, Tsinghua University, Beijing, China;Department of Automation, Tsinghua University, Beijing, China;Department of Automation, Tsinghua University, Beijing, China;Department of Automation, Tsinghua University, Beijing, China
Venue:
MCAM'07 Proceedings of the 2007 international conference on Multimedia content analysis and mining
Year:
2007

Citing 12
Cited 3

The Hierarchical Hidden Markov Model: Analysis and Applications

Machine Learning
A Bayesian Computer Vision System for Modeling Human Interactions

IEEE Transactions on Pattern Analysis and Machine Intelligence
Learning variable-length Markov models of behavior

Computer Vision and Image Understanding - Modeling people toward vision-based underatanding of a person's shape, appearance, and movement
Coupled hidden Markov models for complex action recognition

CVPR '97 Proceedings of the 1997 Conference on Computer Vision and Pattern Recognition (CVPR '97)
Recognition of Group Activities using Dynamic Probabilistic Networks

ICCV '03 Proceedings of the Ninth IEEE International Conference on Computer Vision - Volume 2
Large-Scale Event Detection Using Semi-Hidden Markov Models

ICCV '03 Proceedings of the Ninth IEEE International Conference on Computer Vision - Volume 2
Layered representations for learning and inferring office activity from multiple sensory channels

Computer Vision and Image Understanding - Special issue on event detection in video
Learning and Detecting Activities from Movement Trajectories Using the Hierarchical Hidden Markov Models

CVPR '05 Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 2 - Volume 02
Recognizing Interaction Activities using Dynamic Bayesian Network

ICPR '06 Proceedings of the 18th International Conference on Pattern Recognition - Volume 01
A survey of advances in vision-based human motion capture and analysis

Computer Vision and Image Understanding - Special issue on modeling people: Vision-based understanding of a person's shape, appearance, movement, and behaviour
Policy recognition in the abstract hidden Markov model

Journal of Artificial Intelligence Research
Multi-agent activity recognition using observation decomposedhidden Markov models

Image and Vision Computing

Modeling interleaved hidden processes

Proceedings of the 25th international conference on Machine learning
Hierarchical group process representation in multi-agent activity recognition

Image Communication
Review: The use of pervasive sensing for behaviour profiling - a survey

Pervasive and Mobile Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Dynamic probabilistic networks have been widely used in activity recognition. However, few models are competent for long-term complex activities involving multi-person interactions. Based on the study of activity characteristics, this paper proposes a decomposed hidden Markov model (DHMM) to capture the structures of activity both in time and space. The model combines spatial decomposition and hierarchical abstraction to reduce the complexity of state space as well as the number of parameters greatly, with consequent computational benefits in efficiency and accuracy. Experiments on two-person interactions and individual activities demonstrate that DHMMs are more powerful than Coupled HMMs and Hierarchical HMMs.