venkatesh-humanactivity-2005.pdf (198 kB)
Human activity learning and segmentation using partially hidden discriminative models
conference contribution
posted on 2005-01-01, 00:00 authored by T Truyen, H Bui, Svetha VenkateshSvetha VenkateshLearning and understanding the typical patterns in the daily activities and routines of people from low-level sensory data is an important problem in many application domains such as building smart environments, or providing intelligent assistance. Traditional approaches to this problem typically rely on supervised learning and generative models such as the hidden Markov models and its extensions. While activity data can be readily acquired from pervasive sensors, e.g. in smart environments, providing manual labels to support supervised training is often extremely expensive. In this paper, we propose a new approach based on semi-supervised training of partially hidden discriminative models such as the conditional random field (CRF) and the maximum entropy Markov model (MEMM). We show that these models allow us to incorporate both labeled and unlabeled data for learning, and at the same time, provide us with the flexibility and accuracy of the discriminative framework. Our experimental results in the video surveillance domain illustrate that these models can perform better than their generative counterpart, the partially hidden Markov model, even when a substantial amount of labels are unavailable.
History
Event
International Workshop on Human Activity Recognition and Modelling (2005 : Oxford, U. K.)Pagination
87 - 95Publisher
The Conference, HAREM 2005 in conjunction with BMVC 2005Location
Oxford, U. K.Place of publication
[Oxford, U. K.]Start date
2005-09-09Language
engPublication classification
E1.1 Full written paper - refereedCopyright notice
2005, The AuthorsTitle of proceedings
HAREM 2005 : Proceedings of the International Workshop on Human Activity Recognition and ModellingUsage metrics
Categories
No categories selectedKeywords
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC