An SE(3) invariant description for 3D face recognition

被引:6
|
作者
Jribi, Majdi [1 ]
Rihani, Amal [1 ]
Ben Khlifa, Ameni [1 ]
Ghorbel, Faouzi [1 ]
机构
[1] La Manouba Univ, ENSI, GRIFT Res Grp, CRISTAL Lab, La Manouba 2010, Tunisia
关键词
Multi-polar; Invariance; Face recognition; Geodesic potential; Curvature; Parameterization; KEYPOINT DETECTION; REPRESENTATION; REGISTRATION; DENSE; DEEP;
D O I
10.1016/j.imavis.2019.06.016
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Here, we intend to introduce a novel 3D face description which is invariant under the Special Euclidean group SE(3) and independent to the original surface parameterization. It is well known that it is too difficult to define a relative invariant parameterization of a general curved surface. In the present work, we introduce the multi-polar geodesic representation of R-3 surfaces. It allows to reach an isotropic canonical parameterization relatively to SE(3) due to the fact that the face surfaces can be easily assumed to be a graph of a function from R-2 to R. The principal curvature fields according to a three-polar parameterization are considered. A statistical study is made to choose a good configuration of reference points. The performances of the novel description are tested on the standard database FRGC v2.0. Many recognition scenarios are established. The obtained results are very competitive with the state of the art. (C) 2019 Elsevier B.V. All rights reserved.
引用
收藏
页码:106 / 119
页数:14
相关论文
共 50 条
  • [1] A novel integral and SE(3)-invariant description for 3D face recognition
    Jribi, Majdi
    Othmeni, Zeineb
    Ghorbel, Faouzi
    SIGNAL IMAGE AND VIDEO PROCESSING, 2025, 19 (03)
  • [2] EI3D: Expression-invariant 3D face recognition based on feature and shape matching
    Guo, Yulan
    Lei, Yinjie
    Liu, Li
    Wang, Yan
    Bennamoun, Mohammed
    Sohelf, Ferdous
    PATTERN RECOGNITION LETTERS, 2016, 83 : 403 - 412
  • [3] Face ShapeNets for 3D Face Recognition
    Jabberi, Marwa
    Wali, Ali
    Neji, Bilel
    Beyrouthy, Taha
    Alimi, Adel M.
    IEEE ACCESS, 2023, 11 : 46240 - 46256
  • [4] Robust 3D face recognition based on resolution invariant features
    Zhang, Guangpeng
    Wang, Yunhong
    PATTERN RECOGNITION LETTERS, 2011, 32 (07) : 1009 - 1019
  • [5] IMPROVED 3D ASSISTED POSE-INVARIANT FACE RECOGNITION
    Wang, Liting
    Ding, Liu
    Ding, Xiaoqing
    Fang, Chi
    2009 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOLS 1- 8, PROCEEDINGS, 2009, : 889 - 892
  • [6] Evaluation of implicit 3D modeling for pose invariant face recognition
    Hüsken, M
    Brauckmann, M
    Gehlen, S
    Okada, K
    von der Malsburg, C
    BIOMETRIC TECHNOLOGY FOR HUMAN IDENTIFICATION, 2004, 5404 : 328 - 338
  • [7] A geodesic multipolar parameterization-based representation for 3D face recognition
    Jribi, Majdi
    Mathlouthi, Soumaya
    Ghorbel, Faouzi
    SIGNAL PROCESSING-IMAGE COMMUNICATION, 2021, 99
  • [8] Nasal Patches and Curves for Expression-Robust 3D Face Recognition
    Emambakhsh, Mehryar
    Evans, Adrian
    IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2017, 39 (05) : 995 - 1007
  • [9] 3D FACE REPRESENTATION AND RECOGNITION BY INTRINSIC SHAPE DESCRIPTION MAPS
    Guo, Zhe
    Zhang, Yanning
    Xia, Yong
    Lin, Zenggang
    Feng, Dagan
    2010 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, 2010, : 854 - 857
  • [10] A new strategy for 3D face recognition
    Liu, Xinrong
    Yang, Guang
    Sun, Xiangbin
    Journal of Information and Computational Science, 2015, 12 (07): : 2759 - 2768