Abstract
Human activity recognition is an important and challenging task for video content analysis and understanding. Individual activity recognition has been well studied recently. However, recognizing the activities of human group with more than three people having complex interactions is still a formidable challenge. In this paper, a novel human group activity recognition method is proposed to deal with complex situation where there are multiple sub-groups. To characterize the inherent interactions of intra-subgroups and inter-subgroups with the varying number of participants, this paper proposes three types of group-activity descriptor using motion trajectory and appearance information of people. Experimental results on a public human group activity dataset demonstrate effectiveness of the proposed method.
| Original language | English |
|---|---|
| Title of host publication | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
| Publisher | Springer Verlag |
| Pages | 161-169 |
| Number of pages | 9 |
| Volume | 9315 |
| ISBN (Print) | 9783319240770 |
| DOIs | |
| Publication status | Published - 2015 |
| Event | 16th Pacific-Rim Conference on Multimedia, PCM 2015 - Gwangju, Korea, Republic of Duration: 2015 Sep 16 → 2015 Sep 18 |
Publication series
| Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
|---|---|
| Volume | 9315 |
| ISSN (Print) | 03029743 |
| ISSN (Electronic) | 16113349 |
Other
| Other | 16th Pacific-Rim Conference on Multimedia, PCM 2015 |
|---|---|
| Country | Korea, Republic of |
| City | Gwangju |
| Period | 15/9/16 → 15/9/18 |
Fingerprint
Keywords
- Activity recognition
- Human group activity
- Video analytics
ASJC Scopus subject areas
- Computer Science(all)
- Theoretical Computer Science
Cite this
Recognition of human group activity for video analytics. / Ju, Jaeyong; Yang, Cheoljong; Scherer, Sebastian; Ko, Hanseok.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 9315 Springer Verlag, 2015. p. 161-169 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9315).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Recognition of human group activity for video analytics
AU - Ju, Jaeyong
AU - Yang, Cheoljong
AU - Scherer, Sebastian
AU - Ko, Hanseok
PY - 2015
Y1 - 2015
N2 - Human activity recognition is an important and challenging task for video content analysis and understanding. Individual activity recognition has been well studied recently. However, recognizing the activities of human group with more than three people having complex interactions is still a formidable challenge. In this paper, a novel human group activity recognition method is proposed to deal with complex situation where there are multiple sub-groups. To characterize the inherent interactions of intra-subgroups and inter-subgroups with the varying number of participants, this paper proposes three types of group-activity descriptor using motion trajectory and appearance information of people. Experimental results on a public human group activity dataset demonstrate effectiveness of the proposed method.
AB - Human activity recognition is an important and challenging task for video content analysis and understanding. Individual activity recognition has been well studied recently. However, recognizing the activities of human group with more than three people having complex interactions is still a formidable challenge. In this paper, a novel human group activity recognition method is proposed to deal with complex situation where there are multiple sub-groups. To characterize the inherent interactions of intra-subgroups and inter-subgroups with the varying number of participants, this paper proposes three types of group-activity descriptor using motion trajectory and appearance information of people. Experimental results on a public human group activity dataset demonstrate effectiveness of the proposed method.
KW - Activity recognition
KW - Human group activity
KW - Video analytics
UR - http://www.scopus.com/inward/record.url?scp=84951873430&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84951873430&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-24078-7_16
DO - 10.1007/978-3-319-24078-7_16
M3 - Conference contribution
AN - SCOPUS:84951873430
SN - 9783319240770
VL - 9315
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 161
EP - 169
BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PB - Springer Verlag
ER -