Early features fusion over 3D face for face recognition

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

In this paper, a novel approach for fusing shape and texture Local Binary Patterns (LBP) for 3D Face Recognition is presented. Using the recently proposed mesh-LBP [23], it is now possible to compute LBP directly on a mesh manifold, allowing Early Feature Fusion to enhance face description power. Compared to its depth image counterparts, the proposed method (a) inherits the intrinsic advantages of mesh surfaces, (such as preservation of full geometry), (b) does not require face registration, (c) can accommodate partial or rotation matching, and (d) natively allows early-level fusion of texture and shape descriptors. The advantages of early-fusion is presented together with an experimentation of two merging schemes tested on the Bosphorus database.

Original languageBritish English
Title of host publicationRepresentations, Analysis and Recognition of Shape and Motion from Imaging Data - 6th International Workshop, RFMI 2016, Revised Selected Papers
EditorsFaten Chaieb, Faouzi Ghorbel, Boulbaba Ben Amor
PublisherSpringer Verlag
Pages56-64
Number of pages9
ISBN (Print)9783319606538
DOIs
StatePublished - 2017
Event6th International Workshop on Representations, Analysis and Recognition of Shape and Motion from Imaging Data, RFMI 2016 - Sidi Bou Said Village, Tunisia
Duration: 27 Oct 201629 Oct 2016

Publication series

NameCommunications in Computer and Information Science
Volume684
ISSN (Print)1865-0929

Conference

Conference6th International Workshop on Representations, Analysis and Recognition of Shape and Motion from Imaging Data, RFMI 2016
Country/TerritoryTunisia
CitySidi Bou Said Village
Period27/10/1629/10/16

Keywords

  • 3D face recognition
  • Early feature-fusion
  • LBP
  • Local Binary Pattern

Fingerprint

Dive into the research topics of 'Early features fusion over 3D face for face recognition'. Together they form a unique fingerprint.

Cite this