2015 | OriginalPaper | Chapter
Gender Classification from Face Images Based on Gradient Directional Pattern (GDP)
Authors : Faisal Ahmed, Padma Polash Paul, Patrick Wang, Marina Gavrilova
Published in: Computational Science and Its Applications -- ICCSA 2015
Publisher: Springer International Publishing
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
This paper presents an appearance-based facial feature descriptor based on the gradient directional pattern (GDP) for gender classification from face images. The GDP operator encodes the texture information of a local neighborhood by quantizing the gradient directions of the neighbors with respect to the center. The facial feature descriptor is computed by first dividing the face image into a number of sub-regions and then concatenating the individual GDP histograms computed from the corresponding sub-regions. Then, principal component analysis (PCA) is applied on the obtained face descriptor in order to reduce the feature dimensionality. We use a support vector machine (SVM) for the classification task. Experimental analysis on a large database comprising 1800 facial images shows promising results for the proposed method, as compared to some well-known appearance-based face descriptors.