Gender Classification

Problem Statement

In face recognition, illumination variation has a significant impact on gender classification’s accuracy using visible images.

On the other hand, 3D facial images are not influenced by illumination changes. We want to explore the potential utility of 3D data in face analysis area.

Abstract

Gender classification of depth images is a chal-lenging problem, most research work attempted to use shape information to solve this problem in the past literature. In this work, we propose a new fusion scheme for gender classification using both texture and shape features. A new ensemble scheme is advocated to combine texture and shape feature at the feature level. To evaluate the performance of our algorithm, we measure our scheme on two different datasets. The final classification result is up to 93:7% using five-fold cross validation on the whole FRGCv2 dataset, which is comparable to the classification result obtained using visible imagery.

Illustration of Depth images and their corresponding shape index feature

Related people

X.Wang and C. Kambhamettu

Big View Video/Image Modeling and Synthesis Lab Home
VIMS Home People Research Publications Workshops Achievements Courses Software Data Photos login	Gender Classification Problem Statement In face recognition, illumination variation has a significant impact on gender classification’s accuracy using visible images. On the other hand, 3D facial images are not influenced by illumination changes. We want to explore the potential utility of 3D data in face analysis area. Abstract Gender classification of depth images is a chal-lenging problem, most research work attempted to use shape information to solve this problem in the past literature. In this work, we propose a new fusion scheme for gender classification using both texture and shape features. A new ensemble scheme is advocated to combine texture and shape feature at the feature level. To evaluate the performance of our algorithm, we measure our scheme on two different datasets. The final classification result is up to 93:7% using five-fold cross validation on the whole FRGCv2 dataset, which is comparable to the classification result obtained using visible imagery. Illustration of Depth images and their corresponding shape index feature Related people X.Wang and C. Kambhamettu Page last modified on July 17, 2013, at 02:49 PM