Facial Data Field
Shuliang WangInternational School of SoftwareWuhan UniversityWuhan430079ChinaHanning YuanSchool of SoftwareBeijing Institute of TechnologyBeijing100081ChinaBaohua CaoDepartments of Electrical EngineeringUniversity of Southern CaliforniaLos AngelesCA90089USADakui WangInternational School of SoftwareWuhan UniversityWuhan430079China
ABI
Abstract
Expressional face recognition is a challenge in computer vision for complex expressions. Facial data field is proposed to recognize expression. Fundamentals are presented in the methodology of face recognition upon data field and subsequently, technical algorithms including normalizing faces, generating facial data field, extracting feature points in partitions, assigning weights and recognizing faces. A case is studied with JAFFE database for its verification. Result indicates that the proposed method is suitable and effective in expressional face recognition considering the whole average recognition rate is up to 94.3%. In conclusion, data field is considered as a valuable alternative to pattern recognition.
Topics
Identifiers
Citations and references
Cited by 00 references
Metrics — AkademScholar · Coming soon