Defining mesh-LBP variants for 3D relief patterns classification

Claudio Tortorici, Naoufel Werghi, Stefano Berretti

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

Abstract

Extending the concept of texture to the geometry of a mesh manifold surface, opened the way to the idea of classifying 3D relief patterns as an emerging topic in 3D Computer Vision, with several potential applications. In this paper, we propose an original modelling solution to address this novel task. Following the recent introduction of the LBP computation framework on mesh manifolds (mesh-LBP), we first extend this framework to the different variants of 2D LBP by defining mesh-LBP variants. The compliance of these extensions with the original LBP in terms of uniformity is also investigated. Then, we proposed a complete framework for relief patterns classification, which performs mesh preprocessing, multi-scale mesh-LBP extraction and descriptors classification. Experimental results on the SHREC’17 dataset showed competitive performance with respect to state of the art solutions.

Original languageBritish English
Title of host publicationRepresentations, Analysis and Recognition of Shape and Motion from Imaging Data - 7th International Workshop, RFMI 2017, Revised Selected Papers
EditorsLiming Chen, Faouzi Ghorbel, Boulbaba Ben Amor
PublisherSpringer Verlag
Pages151-166
Number of pages16
ISBN (Print)9783030198152
DOIs
StatePublished - 2019
Event7th International Workshop on Representations, Analysis and Recognition of Shape and Motion from Imaging Data, RFMI 2017 - Savoie, France
Duration: 17 Dec 201720 Dec 2017

Publication series

NameCommunications in Computer and Information Science
Volume842
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937

Conference

Conference7th International Workshop on Representations, Analysis and Recognition of Shape and Motion from Imaging Data, RFMI 2017
Country/TerritoryFrance
CitySavoie
Period17/12/1720/12/17

Fingerprint

Dive into the research topics of 'Defining mesh-LBP variants for 3D relief patterns classification'. Together they form a unique fingerprint.

Cite this