Open AccessPose‐invariant gender classification based on 3D face reconstruction and synthesis from single 2D image
Author(s) -
Moeini A.,
Moeini H.
Publication year - 2015
Publication title -
electronics letters
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.375
H-Index - 146
eISSN - 1350-911X
pISSN - 0013-5194
DOI - 10.1049/el.2015.0520
Subject(s) - invariant (physics) , artificial intelligence , computer vision , face (sociological concept) , facial recognition system , pattern recognition (psychology) , computer science , mathematics , sociology , social science , mathematical physics
A novel method is proposed for pose‐invariant gender classification based on three‐dimensional (3D) face reconstruction from only 2D frontal images. A 3D face model is reconstructed from only a single 2D frontal image. Then, for each two‐class of gender in the database, a feature library matrix (FLM) is created from yaw face poses by rotating the 3D reconstructed models and extracting features in the rotated face. Each FLM is subsequently rendered based on the yaw angles of face poses. Then, an array of the FLM is selected based on the estimated yaw angles for each class of gender. Finally, the selected arrays from FLMs are compared with target image features by support vector machine classification. Promising results are acquired to handle pose in gender classification on the available compared with the state‐of‐the‐art methods.