Using Hidden Markov Models for recognizing action primitives in complex actions
(2007) 15th Scandinavian Conference on Image Analysis, SCIA 2007 In Lecture Notes in Computer Science 4522. p.203-212- Abstract
There is biological evidence that human actions are composed out of action primitives, like words and sentences being composed out of phonemes. Similarly to language processing, one possibility to model and recognize complex actions is to use grammars with action primitives as the alphabet. A major challenge here is that the action primitives need to be recovered first from the noisy input signal before further processing with the action grammar can be done. In this paper we combine a Hidden Markov Model-based approach with a simplified version of a condensation algorithm which allows to recover the action primitives in an observed action. In our approach, the primitives may have different lengths, no clear "divider" between the... (More)
There is biological evidence that human actions are composed out of action primitives, like words and sentences being composed out of phonemes. Similarly to language processing, one possibility to model and recognize complex actions is to use grammars with action primitives as the alphabet. A major challenge here is that the action primitives need to be recovered first from the noisy input signal before further processing with the action grammar can be done. In this paper we combine a Hidden Markov Model-based approach with a simplified version of a condensation algorithm which allows to recover the action primitives in an observed action. In our approach, the primitives may have different lengths, no clear "divider" between the primitives is necessary. The primitive detection is done online, no storing of past data is required. We verify our approach on a large database. Recognition rates are slightly lower than the rate when recognizing the singular action primitives.
(Less)
- author
- Krüger, Volker
LU
and Grest, Daniel
- publishing date
- 2007-12-01
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- host publication
- Image Analysis - 15th Scandinavian Conference, SCIA 2007, Proceedings
- series title
- Lecture Notes in Computer Science
- volume
- 4522
- pages
- 10 pages
- conference name
- 15th Scandinavian Conference on Image Analysis, SCIA 2007
- conference location
- Aalborg, Denmark
- conference dates
- 2007-06-10 - 2007-06-14
- external identifiers
-
- scopus:38049073867
- ISSN
- 1611-3349
- 0302-9743
- ISBN
- 9783540730392
- DOI
- 10.1007/978-3-540-73040-8_21
- language
- English
- LU publication?
- no
- id
- 4c73a69f-4eca-40cb-9d42-448b0cf36208
- date added to LUP
- 2019-07-08 21:17:55
- date last changed
- 2025-04-04 15:22:34
@inproceedings{4c73a69f-4eca-40cb-9d42-448b0cf36208, abstract = {{<p>There is biological evidence that human actions are composed out of action primitives, like words and sentences being composed out of phonemes. Similarly to language processing, one possibility to model and recognize complex actions is to use grammars with action primitives as the alphabet. A major challenge here is that the action primitives need to be recovered first from the noisy input signal before further processing with the action grammar can be done. In this paper we combine a Hidden Markov Model-based approach with a simplified version of a condensation algorithm which allows to recover the action primitives in an observed action. In our approach, the primitives may have different lengths, no clear "divider" between the primitives is necessary. The primitive detection is done online, no storing of past data is required. We verify our approach on a large database. Recognition rates are slightly lower than the rate when recognizing the singular action primitives.</p>}}, author = {{Krüger, Volker and Grest, Daniel}}, booktitle = {{Image Analysis - 15th Scandinavian Conference, SCIA 2007, Proceedings}}, isbn = {{9783540730392}}, issn = {{1611-3349}}, language = {{eng}}, month = {{12}}, pages = {{203--212}}, series = {{Lecture Notes in Computer Science}}, title = {{Using Hidden Markov Models for recognizing action primitives in complex actions}}, url = {{http://dx.doi.org/10.1007/978-3-540-73040-8_21}}, doi = {{10.1007/978-3-540-73040-8_21}}, volume = {{4522}}, year = {{2007}}, }