Title :
Viewpoint-specific active appearance model for robust feature point extraction
Author :
Kawarazaki, Yasutomo ; Duan, Guifang ; Shinkawa, Takuya ; Chen, Yen-Wei
Author_Institution :
Grad. Sch. of Sci. & Eng., Ritsumeikan Univ., Kusatsu, Japan
fDate :
Nov. 29 2011-Dec. 1 2011
Abstract :
Active Appearance Model (AAM) is a popular technique for facial feature point extraction. However, conventional AAM is always viewpoint-generic, which lowers the fitting accuracy. In this paper, a viewpoint-specific AMM is proposed for robust facial point extraction under multi-viewpoint. We investigate three viewpoints, such as 0 degrees, 30 degrees, 60 degrees. The experimental results show a briefly results. By the proposed method, the actuary of the facial point extraction can be largely improved.
Keywords :
face recognition; feature extraction; facial feature point extraction; multiviewpoint; robust feature point extraction; viewpoint-specific AMM; viewpoint-specific active appearance model; Active appearance model; Equations; Face; Feature extraction; Fitting; Mathematical model; Shape; Active Appearance Model; Feature point; Viewpoint-specific;
Conference_Titel :
Computer Sciences and Convergence Information Technology (ICCIT), 2011 6th International Conference on
Conference_Location :
Seogwipo
Print_ISBN :
978-1-4577-0472-7
