Abstract
Previous research on automatic laughter detection has mainly been focused on audio-based detection. In
this study we present an audiovisual approach to distinguishing laughter from speech based on temporal
features and we show that the integration of audio and visual information leads to improved performance
over single-modal approaches. Static features are extracted on an audio/video frame basis and then combined
with temporal features extracted over a temporal window, describing the evolution of static features
over time. When tested on 96 audiovisual sequences, depicting spontaneously displayed (as opposed to
posed) laughter and speech episodes, in a person independent way the proposed audiovisual approach
achieves an F1 rate of over 89%.
Original language | English |
---|---|
Title of host publication | BNAIC 2008 |
Subtitle of host publication | Proceedings of BNAIC 2008, the twentieth Belgian-Dutch Artificial Intelligence Conference, Enschede/Bad Boekelo, October 30-31, 2008 |
Editors | Anton Nijholt, Maja Pantic, Mannes Poel, Hendri Hondorp |
Publisher | University of Twente |
Pages | 351-352 |
Number of pages | 2 |
Publication status | Published - 30 Oct 2008 |
Event | 20th Benelux Conference on Artificial Intelligence, BNAIC 2008 - Boekelo, Netherlands Duration: 30 Oct 2008 → 31 Oct 2008 Conference number: 20 |
Publication series
Name | BNAIC: proceedings of the ... Belgium/Netherlands Artificial Intelligence Conference |
---|---|
Publisher | University of Twente |
Number | 20 |
ISSN (Print) | 1568-7805 |
Conference
Conference | 20th Benelux Conference on Artificial Intelligence, BNAIC 2008 |
---|---|
Abbreviated title | BNAIC |
Country | Netherlands |
City | Boekelo |
Period | 30/10/08 → 31/10/08 |
Keywords
- HMI-MI: MULTIMODAL INTERACTIONS
- EC Grant Agreement nr.: FP7/211486
- EC Grant Agreement nr.: FP6/0027787