Accumulating weighted segmentation in 3D face recognition. (24th April 2022)

• Record Type: Journal Article
• Authors: Ju, Quan; Hu, Haitao; Wang, Yingfeng
• Is Part Of: International journal of wireless and mobile computing. Volume 22:Number 1(2022)
• Page Start: 84
• Page End: 92
• Publication Date: 2022-04-24
• Subjects: face segmentation -- expression variation -- accumulating weight -- face recognition in 3D; Mobile computing -- Periodicals; Wireless communication systems -- Periodicals; 004.6
• Languages: English
• ISSNs: 1741-1084

Abstract

In this paper, an accumulating weighted face segmentation approach based on the rigid level of human facial areas is introduced. A mass of 3D face data is measured and analysed to define the most expression-invariant region. Different locations or regions on the human face are observed to have dissimilar invariant levels. Thus, an accumulating weight method is proposed to represent the rigid degree under expression variations. In face identification experiments, performance by employing the accumulating weight is demonstrated to be higher than methods using the expression-invariant region and the full face, respectively. This accumulating weighted face segmentation approach outperforms other state-of-the-art methods in 3D face recognition experiments.