Classifying head movements in video-recorded conversations based on movement velocity, acceleration and jerk

Research output: Contribution to journalConference articlepeer-review

This paper is about the automatic annotation of head movements in videos of face-to-face conversations. Manual annotation of gestures is resource consuming, and modelling gesture behaviours in different types of communicative settings requires many types of annotated data. Therefore, developing methods for automatic annotation is crucial. We present an approach where an SVM classifier learns to classify head movements based on measurements of velocity, acceleration, and the third derivative of position with respect to time, jerk. Consequently, annotations of head movements are added to new video data. The results of the automatic annotation are evaluated against manual annotations in the same data and show an accuracy of 73.47% with respect to these. The results also show that using jerk improves accuracy.
Original languageEnglish
Article number003
JournalLinköping Electronic Conference Proceedings
Issue number141
Pages (from-to)10-17
ISSN1650-3740
Publication statusPublished - 2017
EventNordic and European Symposium on Multimodal Communication : 7th Nordic and 4th European Symposium on Multimodal Communication - University of Copenhagen, Copenhagen, Denmark
Duration: 29 Sep 201630 Sep 2016
Conference number: 4th, 7th
http://mmsym.org/?page_id=412

Conference

ConferenceNordic and European Symposium on Multimodal Communication
Number4th, 7th
LocationUniversity of Copenhagen
CountryDenmark
CityCopenhagen
Period29/09/201630/09/2016
Internet address

ID: 183642602