Human action recognition using time delay input radial basis function networks
This paper presents a fast, vision-based method for the problem of human action representation and recognition. The first problem is addressed by constructing an action descriptor from spatiotemporal data of action silhouettes based on appearance and motion features. For action classification, a new...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
United Kingdom Simulation Society
2014
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/37653/ http://psasir.upm.edu.my/id/eprint/37653/1/Human%20action%20recognition%20using%20time%20delay%20input%20radial%20basis%20function%20networks.pdf |
| _version_ | 1848848665919094784 |
|---|---|
| author | Kalhor, Davood Aris, Ishak Moaini, Trifa Abdul Halin, Izhal |
| author_facet | Kalhor, Davood Aris, Ishak Moaini, Trifa Abdul Halin, Izhal |
| author_sort | Kalhor, Davood |
| building | UPM Institutional Repository |
| collection | Online Access |
| description | This paper presents a fast, vision-based method for the problem of human action representation and recognition. The first problem is addressed by constructing an action descriptor from spatiotemporal data of action silhouettes based on appearance and motion features. For action classification, a new Radial Basis Function Network (RBF), called Time Delay Input Radial Basis Function Network (TDIRBF) is proposed by introducing time delay units to the RBF in a novel approach. A TDIRBF offers a few desirable features such as an easier learning process and more flexibility. The representational power and speed of the proposed method were explored using a publicly available dataset. Based on experimental results, implemented in MATLAB and on standard PCs, the average time for constructing a feature vector for a high-resolution video was just about 20 ms/frame (or 50 fps) and the classifier speed was above 15 fps. Furthermore, the proposed approach demonstrated good performance in terms of both execution time and overall performance (a new performance measure that combines accuracy and speed into one metric). |
| first_indexed | 2025-11-15T09:38:07Z |
| format | Article |
| id | upm-37653 |
| institution | Universiti Putra Malaysia |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-15T09:38:07Z |
| publishDate | 2014 |
| publisher | United Kingdom Simulation Society |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | upm-376532017-10-30T06:11:03Z http://psasir.upm.edu.my/id/eprint/37653/ Human action recognition using time delay input radial basis function networks Kalhor, Davood Aris, Ishak Moaini, Trifa Abdul Halin, Izhal This paper presents a fast, vision-based method for the problem of human action representation and recognition. The first problem is addressed by constructing an action descriptor from spatiotemporal data of action silhouettes based on appearance and motion features. For action classification, a new Radial Basis Function Network (RBF), called Time Delay Input Radial Basis Function Network (TDIRBF) is proposed by introducing time delay units to the RBF in a novel approach. A TDIRBF offers a few desirable features such as an easier learning process and more flexibility. The representational power and speed of the proposed method were explored using a publicly available dataset. Based on experimental results, implemented in MATLAB and on standard PCs, the average time for constructing a feature vector for a high-resolution video was just about 20 ms/frame (or 50 fps) and the classifier speed was above 15 fps. Furthermore, the proposed approach demonstrated good performance in terms of both execution time and overall performance (a new performance measure that combines accuracy and speed into one metric). United Kingdom Simulation Society 2014 Article PeerReviewed application/pdf en http://psasir.upm.edu.my/id/eprint/37653/1/Human%20action%20recognition%20using%20time%20delay%20input%20radial%20basis%20function%20networks.pdf Kalhor, Davood and Aris, Ishak and Moaini, Trifa and Abdul Halin, Izhal (2014) Human action recognition using time delay input radial basis function networks. International Journal of Simulation: Systems, Science & Technology, 15 (3). pp. 42-53. ISSN 1473-8031; ESSN: 1473-804X http://ijssst.info/Vol-15/No-3/cover-15-3.htm 10.5013/IJSSST.a.15.03.07 |
| spellingShingle | Kalhor, Davood Aris, Ishak Moaini, Trifa Abdul Halin, Izhal Human action recognition using time delay input radial basis function networks |
| title | Human action recognition using time delay input radial basis function networks |
| title_full | Human action recognition using time delay input radial basis function networks |
| title_fullStr | Human action recognition using time delay input radial basis function networks |
| title_full_unstemmed | Human action recognition using time delay input radial basis function networks |
| title_short | Human action recognition using time delay input radial basis function networks |
| title_sort | human action recognition using time delay input radial basis function networks |
| url | http://psasir.upm.edu.my/id/eprint/37653/ http://psasir.upm.edu.my/id/eprint/37653/ http://psasir.upm.edu.my/id/eprint/37653/ http://psasir.upm.edu.my/id/eprint/37653/1/Human%20action%20recognition%20using%20time%20delay%20input%20radial%20basis%20function%20networks.pdf |