Date of Graduation


Document Type


Degree Type



Statler College of Engineering and Mineral Resources


Lane Department of Computer Science and Electrical Engineering

Committee Chair

Arun Ross.


Gender classification based on facial images has received increased attention in the computer vision community. In this work, a comprehensive evaluation of state-of-the-art gender classification methods is carried out on publicly available databases and extended to reallife face images, where face detection and face normalization are essential for the success of the system. Next, the possibility of predicting gender from face images acquired in the near-infrared spectrum (NIR) is explored. In this regard, the following two questions are addressed: (a) Can gender be predicted from NIR face images; and (b) Can a gender predictor learned using visible (VIS) images operate successfully on NIR images and vice-versa? The experimental results suggest that NIR face images do have some discriminatory information pertaining to gender, although the degree of discrimination is noticeably lower than that of VIS images. Further, the use of an illumination normalization routine may be essential for facilitating cross-spectral gender prediction. By formulating the problem of gender classification in the framework of both visible and near-infrared images, the guidelines for performing gender classification in a real-world scenario is provided, along with the strengths and weaknesses of each methodology. Finally, the general problem of attribute classification is addressed, where features such as expression, age and ethnicity are derived from a face image.