On the use of Kinect depth data for identity, gender and ethnicity classification from facial images

• Investigate the usefulness of the depth images provided by RGB-D sensors in face analysis tasks.
• Provide a comprehensive review of the prior work and present publicly available RGB-D face databases.
• Compare the performance of the depth images against RGB counterpart in three face analysis tasks.

This article explores the usefulness of the depth images provided by the current Microsoft Kinect sensors in different face analysis tasks including identity, gender and ethnicity. Four local feature extraction methods (LBP, LPQ, HOG and BSIF) are investigated for both face texture and shape description. Extensive experiments on three publicly available Kinect face databases are reported. The experimental analysis yields into interesting findings. Furthermore, a comprehensive review of the literature on the use of Kinect depth data in face analysis is provided along with the description of the available databases.

Figure optionsDownload high-quality image (160 K)Download as PowerPoint slide