Gender classification from 3D face images using multi-task sparse representation over reduced

S. Bentaieb , A. Ouamri , M. Keche , A. A. Nait-Ali (1)

1 SYNAPSE

In this paper, we address the problem of gender classification based on facial images. The Speeded Up Robust Feature (SURF) algorithm descriptors are used as features to built dictionaries and a multi-task Sparse Representation Classification (SRC) is used as classifier to determine the gender of an individual face. Our approach uses smaller and compact dictionaries by removing the redundant atoms from the constructed ones. The feasibility of using the SURF on the shape index map for gender classification is demonstrated through experimental investigation conducted on FRGCv2 dataset. The proposed approach achieves 91.04±1.19% of correct gender classification rate using only 5% of the size of the dictionary and 97.83 ± 0.76% is obtained using 23% of the size of the dictionary.

Mots clés: Dictionaries; Face; Three-dimensional displays; Shape; Indexes; Training

Domaines: Informatique [cs]

S. Bentaieb, A. Ouamri, M. Keche, A. A. Nait-Ali. Gender classification from 3D face images using multi-task sparse representation over reduced. 2nd International Conference on Bio-engineering for Smart Technologies (BioSMART), Aug 2017, Paris, France. ⟨10.1109/BIOSMART.2017.8095327⟩. ⟨hal-01865328⟩