• DocumentCode
    160299
  • Title

    Chinese Facial Expression Database with spontaneous expressions across multiple poses

  • Author

    Xiaoxue Cheng ; Lixia Xue ; Yanxiang Chen ; Lei Liu

  • Author_Institution
    Sch. of Comput. Sci. & Inf., Hefei Univ. of Technol., Hefei, China
  • fYear
    2014
  • fDate
    11-13 July 2014
  • Firstpage
    1
  • Lastpage
    6
  • Abstract
    Although exist several European and American databases dealing with facial expressions, there are no equivalent databases for Asian facial expressions which have definitely different characteristics from those of western people. Besides database KFDB collecting for Korean subjects and JAFFE composed of Japanese female facial expressions, there are few Chinese facial expression databases. Chinese database CAS-PEAL consists only of posed expressions, whereas NVIE focuses only on frontal expressions. Therefore, to enrich the existing databases, we propose and establish a new one for Chinese subjects named Chinese Facial Expression Database (CFED). In this database, spontaneous expressions besides posed expressions are obtained across multiple poses besides frontal pose. Furthermore, the database is annotated with description files for exact capture environments and ground truth information for facial feature points.
  • Keywords
    face recognition; feature extraction; visual databases; American database; Asian facial expression; CFED database; Chinese facial expression database; European database; JAFFE database; KFDB database; database annotation; exact capture environment; facial feature points; frontal expression; ground truth information; Decision support systems; Chinese facial expression database; Chinese subjects; Spontaneous expression; across different poses; facial feature points;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computing, Communication and Networking Technologies (ICCCNT), 2014 International Conference on
  • Conference_Location
    Hefei
  • Print_ISBN
    978-1-4799-2695-4
  • Type

    conf

  • DOI
    10.1109/ICCCNT.2014.6963009
  • Filename
    6963009