The Influence of Facial Width-to-Height Ratio on Micro-expression Recognition - Intelligence Science II Third IFIP TC 12 International Conference, ICIS 2018 Accéder directement au contenu
Communication Dans Un Congrès Année : 2018

The Influence of Facial Width-to-Height Ratio on Micro-expression Recognition

Qi Wu
  • Fonction : Auteur
  • PersonId : 1046642

Résumé

The aim of the present study was to uncover the potential impact of facial width-to-height ratio (fWHR) on micro-expression and macro-expression recognition. The JACBART paradigm was used for the presentation of facial expressions. Participants were asked to recognize six kinds of basic expressions (sadness, surprise, anger, disgust, fear, happiness) on high fWHR faces or on low fWHR faces under 67 ms, 333 ms and 600 ms duration conditions respectively. The results indicated that, the fWHR did not affect the recognition of macro-expressions which were presented for 600 ms in the present study, but the fWHR could influence the recognition accuracy of micro-expressions of surprise and happiness. Specifically, participants could identify the facial expression of surprise more effectively on high fWHR faces than on low fWHR faces under the condition of 67 ms. And participants also could recognize the facial expression of happiness more accurately on high fWHR faces in the conditions of 67 ms and 333 ms. These results revealed the facial expressions of happiness and surprise on high fWHR faces may have an early processing advantage in micro-expression recognition. And the result also demonstrated that individuals spontaneously use fWHR as a clue to recognize micro-expressions.
Fichier principal
Vignette du fichier
474230_1_En_28_Chapter.pdf (352.39 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02118823 , version 1 (03-05-2019)

Licence

Paternité

Identifiants

Citer

Siwei Zhang, Jinyuan Xie, Qi Wu. The Influence of Facial Width-to-Height Ratio on Micro-expression Recognition. 2nd International Conference on Intelligence Science (ICIS), Nov 2018, Beijing, China. pp.265-272, ⟨10.1007/978-3-030-01313-4_28⟩. ⟨hal-02118823⟩
72 Consultations
148 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More