• DocumentCode
    2368108
  • Title

    Evaluation study on asymmetrical facial expressions generation for Humanoid Robot

  • Author

    Trovato, G. ; Kishi, T. ; Endo, N. ; Hashimoto, Koji ; Takanishi, A.

  • Author_Institution
    Grad. Sch. of Sci. & Eng., Waseda Univ., Tokyo, Japan
  • fYear
    2012
  • fDate
    7-9 Dec. 2012
  • Firstpage
    129
  • Lastpage
    134
  • Abstract
    Human-Robot Interaction is one of the biggest challenges in Humanoid Robotics. Interaction can be carried out through different channels, as humans can communicate through complex languages, including the use of non-verbal communication. Specifically, facial expressions are important for conveying emotions and communication intentions. Several humanoid robots can already perform a certain set of expressions, but their capabilities are somewhat limited and as a result, interaction is not natural. One aspect that could help robots being perceived as real is asymmetry. That is why in this paper we perform an evaluation study of symmetrical and asymmetrical facial expressions of the humanoid robot KOBIAN-R These expressions are made by a generator based on relevant studies of facial anatomyand non-verbal communication and on the work of illustrators and cartoonists. Survey results confirmed the effectiveness of the generator and the importance of asymmetry. We conclude that using this system, robot communication capabilities improve, making possible the development of more advanced interaction in the future.
  • Keywords
    emotion recognition; human-robot interaction; humanoid robots; KOBIAN-R humanoid robot; asymmetrical facial expressions generation; cartoonist; evaluation study; facial anatomy; human-robot interaction; illustrator; nonverbal communication; symmetrical facial expression; Emotion Recognition; Facial Expressions; Human-Robot Interaction;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Innovative Engineering Systems (ICIES), 2012 First International Conference on
  • Conference_Location
    Alexandria
  • Print_ISBN
    978-1-4673-4440-1
  • Type

    conf

  • DOI
    10.1109/ICIES.2012.6530858
  • Filename
    6530858