Abstract
A human action recognition method based on a hidden Markov model (HMM) is proposed. It is a feature-based bottom-up approach that is characterized by its learning capability and time-scale invariability. To apply HMMs, one set of time-sequential images is transformed into an image feature vector sequence, and the sequence is converted into a symbol sequence by vector quantization. In learning human action categories, the parameters of the HMMs, one per category, are optimized so as to best describe the training sequences from the category. To recognize an observed sequence, the HMM which best matches the sequence is chosen. Experimental results for real time-sequential images of sports scenes show recognition rates higher than 90%. The recognition rate is improved by increasing the number of people used to generate the training data, indicating the possibility of establishing a person-independent action recognizer.
Original language | English |
---|---|
Title of host publication | Proceedings CVPR 1992 - IEEE Computer Society Conference on Computer Vision and Pattern Recognition |
Publisher | IEEE Computer Society |
Pages | 379-385 |
Number of pages | 7 |
ISBN (Electronic) | 0818628553 |
DOIs | |
Publication status | Published - 1992 Jan 1 |
Externally published | Yes |
Event | 1992 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 1992 - Champaign, United States Duration: 1992 Jun 15 → 1992 Jun 18 |
Publication series
Name | Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition |
---|---|
Volume | 1992-June |
ISSN (Print) | 1063-6919 |
Conference
Conference | 1992 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 1992 |
---|---|
Country | United States |
City | Champaign |
Period | 92/6/15 → 92/6/18 |
Fingerprint
ASJC Scopus subject areas
- Software
- Computer Vision and Pattern Recognition
Cite this
Recognizing human action in time-sequential images using hidden Markov model. / Yamato, J.; Ohya, Jun; Ishii, K.
Proceedings CVPR 1992 - IEEE Computer Society Conference on Computer Vision and Pattern Recognition. IEEE Computer Society, 1992. p. 379-385 223161 (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition; Vol. 1992-June).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Recognizing human action in time-sequential images using hidden Markov model
AU - Yamato, J.
AU - Ohya, Jun
AU - Ishii, K.
PY - 1992/1/1
Y1 - 1992/1/1
N2 - A human action recognition method based on a hidden Markov model (HMM) is proposed. It is a feature-based bottom-up approach that is characterized by its learning capability and time-scale invariability. To apply HMMs, one set of time-sequential images is transformed into an image feature vector sequence, and the sequence is converted into a symbol sequence by vector quantization. In learning human action categories, the parameters of the HMMs, one per category, are optimized so as to best describe the training sequences from the category. To recognize an observed sequence, the HMM which best matches the sequence is chosen. Experimental results for real time-sequential images of sports scenes show recognition rates higher than 90%. The recognition rate is improved by increasing the number of people used to generate the training data, indicating the possibility of establishing a person-independent action recognizer.
AB - A human action recognition method based on a hidden Markov model (HMM) is proposed. It is a feature-based bottom-up approach that is characterized by its learning capability and time-scale invariability. To apply HMMs, one set of time-sequential images is transformed into an image feature vector sequence, and the sequence is converted into a symbol sequence by vector quantization. In learning human action categories, the parameters of the HMMs, one per category, are optimized so as to best describe the training sequences from the category. To recognize an observed sequence, the HMM which best matches the sequence is chosen. Experimental results for real time-sequential images of sports scenes show recognition rates higher than 90%. The recognition rate is improved by increasing the number of people used to generate the training data, indicating the possibility of establishing a person-independent action recognizer.
UR - http://www.scopus.com/inward/record.url?scp=85060905667&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85060905667&partnerID=8YFLogxK
U2 - 10.1109/CVPR.1992.223161
DO - 10.1109/CVPR.1992.223161
M3 - Conference contribution
AN - SCOPUS:85060905667
T3 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
SP - 379
EP - 385
BT - Proceedings CVPR 1992 - IEEE Computer Society Conference on Computer Vision and Pattern Recognition
PB - IEEE Computer Society
ER -