Automatic Detection and Classification of Head Movements in Face-to-Face Conversations
Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › fagfællebedømt
Standard
Automatic Detection and Classification of Head Movements in Face-to-Face Conversations. / Paggio, Patrizia; Aguirrezabal Zabaleta, Manex; Jongejan, Bart; Navarretta, Costanza.
Proceedings of LREC2020 Workshop "People in language, vision and the mind'' (ONION2020). European Language Resources Association, 2020. s. 15-21.Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › fagfællebedømt
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - GEN
T1 - Automatic Detection and Classification of Head Movements in Face-to-Face Conversations
AU - Paggio, Patrizia
AU - Aguirrezabal Zabaleta, Manex
AU - Jongejan, Bart
AU - Navarretta, Costanza
PY - 2020
Y1 - 2020
N2 - This paper presents an approach to automatic head movement detection and classification in data from a corpus of video-recorded face-to-face conversations in Danish involving 12 different speakers. A number of classifiers were trained with different combinations of visual, acoustic and word features and tested in a leave-one-out cross validation scenario. The visual movement features were extracted from the raw video data using OpenPose, and the acoustic ones using Praat. The best results were obtained by a Multilayer Perceptron classifier, which reached an average 0.68 F1 score across the 12 speakers for head movement detection, and 0.40 for head movement classification given four different classes. In both cases, the classifier outperformed a simple most frequent class baseline as well as a more advanced baseline only relying on velocity features.
AB - This paper presents an approach to automatic head movement detection and classification in data from a corpus of video-recorded face-to-face conversations in Danish involving 12 different speakers. A number of classifiers were trained with different combinations of visual, acoustic and word features and tested in a leave-one-out cross validation scenario. The visual movement features were extracted from the raw video data using OpenPose, and the acoustic ones using Praat. The best results were obtained by a Multilayer Perceptron classifier, which reached an average 0.68 F1 score across the 12 speakers for head movement detection, and 0.40 for head movement classification given four different classes. In both cases, the classifier outperformed a simple most frequent class baseline as well as a more advanced baseline only relying on velocity features.
M3 - Article in proceedings
SP - 15
EP - 21
BT - Proceedings of LREC2020 Workshop "People in language, vision and the mind'' (ONION2020)
PB - European Language Resources Association
ER -
ID: 243519048