Head Pose Estimation Using Simple Local Gabor Binary Pattern

Download PDF
Open Webpage

Weijun Hu, Bingpeng Ma, Xiujuan Chai

Published: 2011, Last Modified: 28 Feb 2026CCBR 2011EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: In this paper, a novel method named simple Local Gabor Binary Pattern method (sLGBP) is presented to improve the accuracy of head pose estimation. The motivation of sLGBP comes from the great success of Local Gabor Binary Pattern (LGBP) in many areas. Considering the relationship between the symmetry of face image and the head pose, the Gabor filters and the LBP operators in sLGBP are all based on the one-dimension. By this way, feature extracted by sLGBP is more related to the head pose while the compute efficiency is improved greatly. To show the effectiveness of sLGBP, we compared them with other methods under two different databases. The results of the experiments show that the proposed methods can improve the accuracy of head pose estimation.