Human action recognition in table-top scenarios : An HMM-based analysis to optimize the performance
(2007) 12th International Conference on Computer Analysis of Images and Patterns, CAIP 2007 In Lecture Notes in Computer Science 4673. p.101-108- Abstract
Hidden Markov models have been extensively and successfully used for the recognition of human actions. Though there exist wellestablished algorithms to optimize the transition and output probabilities, the type of features to use and specifically the number of states and Gaussian have to be chosen manually. Here we present a quantitative study on selecting the optimal feature set for recognition of simple object manipulation actions pointing, rotating and grasping in a tabletop scenario. This study has resulted in recognition rate higher than 90%. Also three different parameters, namely the number of states and Gaussian for HMM and the number of training iterations, are considered for optimization of the recognition rate with 5... (More)
Hidden Markov models have been extensively and successfully used for the recognition of human actions. Though there exist wellestablished algorithms to optimize the transition and output probabilities, the type of features to use and specifically the number of states and Gaussian have to be chosen manually. Here we present a quantitative study on selecting the optimal feature set for recognition of simple object manipulation actions pointing, rotating and grasping in a tabletop scenario. This study has resulted in recognition rate higher than 90%. Also three different parameters, namely the number of states and Gaussian for HMM and the number of training iterations, are considered for optimization of the recognition rate with 5 different feature sets on our motion capture data set from 10 persons.
(Less)
- author
- Raamana, Pradeep Reddy ; Grest, Daniel and Krueger, Volker LU
- publishing date
- 2007-12-01
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- Action recognition, Hidden Markov model, Optimization
- host publication
- Computer Analysis of Images and Patterns - 12th International Conference, CAIP 2007, Proceedings
- series title
- Lecture Notes in Computer Science
- volume
- 4673
- pages
- 8 pages
- conference name
- 12th International Conference on Computer Analysis of Images and Patterns, CAIP 2007
- conference location
- Vienna, Austria
- conference dates
- 2007-08-27 - 2007-08-29
- external identifiers
-
- scopus:38349011918
- ISSN
- 1611-3349
- 0302-9743
- ISBN
- 9783540742715
- DOI
- 10.1007/978-3-540-74272-2_13
- language
- English
- LU publication?
- no
- id
- 8875a4e8-2cd6-4d86-980a-dc59d39e2358
- date added to LUP
- 2019-07-08 21:16:39
- date last changed
- 2024-10-02 09:18:10
@inproceedings{8875a4e8-2cd6-4d86-980a-dc59d39e2358, abstract = {{<p>Hidden Markov models have been extensively and successfully used for the recognition of human actions. Though there exist wellestablished algorithms to optimize the transition and output probabilities, the type of features to use and specifically the number of states and Gaussian have to be chosen manually. Here we present a quantitative study on selecting the optimal feature set for recognition of simple object manipulation actions pointing, rotating and grasping in a tabletop scenario. This study has resulted in recognition rate higher than 90%. Also three different parameters, namely the number of states and Gaussian for HMM and the number of training iterations, are considered for optimization of the recognition rate with 5 different feature sets on our motion capture data set from 10 persons.</p>}}, author = {{Raamana, Pradeep Reddy and Grest, Daniel and Krueger, Volker}}, booktitle = {{Computer Analysis of Images and Patterns - 12th International Conference, CAIP 2007, Proceedings}}, isbn = {{9783540742715}}, issn = {{1611-3349}}, keywords = {{Action recognition; Hidden Markov model; Optimization}}, language = {{eng}}, month = {{12}}, pages = {{101--108}}, series = {{Lecture Notes in Computer Science}}, title = {{Human action recognition in table-top scenarios : An HMM-based analysis to optimize the performance}}, url = {{http://dx.doi.org/10.1007/978-3-540-74272-2_13}}, doi = {{10.1007/978-3-540-74272-2_13}}, volume = {{4673}}, year = {{2007}}, }