3D face recognition using local binary patterns
Hengliang Tang,Baocai Yin,Yanfeng Sun,Yongli Hu
DOI: https://doi.org/10.1016/j.sigpro.2012.04.002
IF: 4.729
2013-01-01
Signal Processing
Abstract:It is well recognized that expressions can significantly change facial geometry that results in a severe problem for robust 3D face recognition. So it is crucial for many applications that how to extract expression-robust features to describe 3D faces. In this paper, we develop a novel 3D face recognition algorithm using Local Binary Pattern (LBP) under expression varieties, which is an extension of the LBP operator widely used in ordinary facial analysis. First, to depict the human face more accurately and reduce the effect of facial local distortion for face recognition, a special feature-based 3D face division scheme is proposed. Then, the LBP representation framework for 3D faces is described, and the facial depth and normal information are extracted and encoded by LBP, to reduce the expression effect. For each face region, the statistical histogram is utilized to summarize the facial details, and accordingly three matching strategies are presented to address the recognition task. Finally, the proposed 3D face recognition algorithm is tested on BJUT-3D and FRGC v2.0 databases, achieves promising results, and concludes that it is feasible and valid to apply the LBP representation framework on 3D face recognition.