Material classification on symmetric positive definite manifolds

Masoud Faraki; Mehrtash T. Harandi; Fatih Porikli

doi:10.1109/WACV.2015.105

Material classification on symmetric positive definite manifolds

Masoud Faraki
, Mehrtash T. Harandi
, Fatih Porikli

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

13 Citations (Scopus)

Abstract

This paper tackles the problem of categorizing materials and textures by exploiting the second order statistics. To this end, we introduce the Extrinsic Vector of Locally Aggregated Descriptors (E-VLAD), a method to combine local and structured descriptors into a unified vector representation where each local descriptor is a Covariance Descriptor (CovD). In doing so, we make use of an accelerated method of obtaining a visual codebook where each atom is itself a CovD. We will then introduce an efficient way of aggregating local CovDs into a vector representation. Our method could be understood as an extrinsic extension of the highly acclaimed method of Vector of Locally Aggregated Descriptors [17] (or VLAD) to CovDs. We will show that the proposed method is extremely powerful in classifying materials/ textures and can outperform complex machineries even with simple classifiers.

Original language	English
Title of host publication	2015 IEEE Winter Conference on Applications of Computer Vision (WACV 2015)
Editors	Sudeep Sarkar, Subhodev Das, Bahram Parvin, Fatih Porikli
Place of Publication	Piscataway NJ USA
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	749-756
Number of pages	8
ISBN (Print)	9781479966844
DOIs	https://doi.org/10.1109/WACV.2015.105
Publication status	Published - 1 Jan 2015
Externally published	Yes
Event	IEEE Winter Conference on Applications of Computer Vision 2015 - Waikoloa Beach Marriott Resort & Spa, Big Island, United States of America Duration: 5 Jan 2015 → 9 Jan 2015 http://wacv2015.org/ https://ieeexplore.ieee.org/xpl/conhome/7045624/proceeding (Proceedings)

Conference

Conference	IEEE Winter Conference on Applications of Computer Vision 2015
Abbreviated title	WACV 2015
Country/Territory	United States of America
City	Big Island
Period	5/01/15 → 9/01/15
Internet address	http://wacv2015.org/ https://ieeexplore.ieee.org/xpl/conhome/7045624/proceeding (Proceedings)

Keywords

Material classification
Region covariance descriptor
Riemannian manifolds
Texture recognition
Vector of locally aggregated descriptors

Access to Document

10.1109/WACV.2015.105

Cite this

@inproceedings{c94fe7a1706e4b17ad12522ae4a63f42,

title = "Material classification on symmetric positive definite manifolds",

abstract = "This paper tackles the problem of categorizing materials and textures by exploiting the second order statistics. To this end, we introduce the Extrinsic Vector of Locally Aggregated Descriptors (E-VLAD), a method to combine local and structured descriptors into a unified vector representation where each local descriptor is a Covariance Descriptor (CovD). In doing so, we make use of an accelerated method of obtaining a visual codebook where each atom is itself a CovD. We will then introduce an efficient way of aggregating local CovDs into a vector representation. Our method could be understood as an extrinsic extension of the highly acclaimed method of Vector of Locally Aggregated Descriptors [17] (or VLAD) to CovDs. We will show that the proposed method is extremely powerful in classifying materials/ textures and can outperform complex machineries even with simple classifiers.",

keywords = "Material classification, Region covariance descriptor, Riemannian manifolds, Texture recognition, Vector of locally aggregated descriptors",

author = "Masoud Faraki and Harandi, \{Mehrtash T.\} and Fatih Porikli",

year = "2015",

month = jan,

day = "1",

doi = "10.1109/WACV.2015.105",

language = "English",

isbn = "9781479966844",

pages = "749--756",

editor = "Sudeep Sarkar and Subhodev Das and Bahram Parvin and Fatih Porikli",

booktitle = "2015 IEEE Winter Conference on Applications of Computer Vision (WACV 2015)",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

address = "United States of America",

note = "IEEE Winter Conference on Applications of Computer Vision 2015, WACV 2015 ; Conference date: 05-01-2015 Through 09-01-2015",

url = "http://wacv2015.org/, https://ieeexplore.ieee.org/xpl/conhome/7045624/proceeding",

}

Faraki, M, Harandi, MT & Porikli, F 2015, Material classification on symmetric positive definite manifolds. in S Sarkar, S Das, B Parvin & F Porikli (eds), 2015 IEEE Winter Conference on Applications of Computer Vision (WACV 2015). IEEE, Institute of Electrical and Electronics Engineers, Piscataway NJ USA, pp. 749-756, IEEE Winter Conference on Applications of Computer Vision 2015, Big Island, Hawaii, United States of America, 5/01/15. https://doi.org/10.1109/WACV.2015.105

Material classification on symmetric positive definite manifolds. / Faraki, Masoud; Harandi, Mehrtash T.; Porikli, Fatih.
2015 IEEE Winter Conference on Applications of Computer Vision (WACV 2015). ed. / Sudeep Sarkar; Subhodev Das; Bahram Parvin; Fatih Porikli. Piscataway NJ USA: IEEE, Institute of Electrical and Electronics Engineers, 2015. p. 749-756.

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - Material classification on symmetric positive definite manifolds

AU - Faraki, Masoud

AU - Harandi, Mehrtash T.

AU - Porikli, Fatih

PY - 2015/1/1

Y1 - 2015/1/1

N2 - This paper tackles the problem of categorizing materials and textures by exploiting the second order statistics. To this end, we introduce the Extrinsic Vector of Locally Aggregated Descriptors (E-VLAD), a method to combine local and structured descriptors into a unified vector representation where each local descriptor is a Covariance Descriptor (CovD). In doing so, we make use of an accelerated method of obtaining a visual codebook where each atom is itself a CovD. We will then introduce an efficient way of aggregating local CovDs into a vector representation. Our method could be understood as an extrinsic extension of the highly acclaimed method of Vector of Locally Aggregated Descriptors [17] (or VLAD) to CovDs. We will show that the proposed method is extremely powerful in classifying materials/ textures and can outperform complex machineries even with simple classifiers.

AB - This paper tackles the problem of categorizing materials and textures by exploiting the second order statistics. To this end, we introduce the Extrinsic Vector of Locally Aggregated Descriptors (E-VLAD), a method to combine local and structured descriptors into a unified vector representation where each local descriptor is a Covariance Descriptor (CovD). In doing so, we make use of an accelerated method of obtaining a visual codebook where each atom is itself a CovD. We will then introduce an efficient way of aggregating local CovDs into a vector representation. Our method could be understood as an extrinsic extension of the highly acclaimed method of Vector of Locally Aggregated Descriptors [17] (or VLAD) to CovDs. We will show that the proposed method is extremely powerful in classifying materials/ textures and can outperform complex machineries even with simple classifiers.

KW - Material classification

KW - Region covariance descriptor

KW - Riemannian manifolds

KW - Texture recognition

KW - Vector of locally aggregated descriptors

UR - https://www.scopus.com/pages/publications/84925424789

U2 - 10.1109/WACV.2015.105

DO - 10.1109/WACV.2015.105

M3 - Conference Paper

AN - SCOPUS:84925424789

SN - 9781479966844

SP - 749

EP - 756

BT - 2015 IEEE Winter Conference on Applications of Computer Vision (WACV 2015)

A2 - Sarkar, Sudeep

A2 - Das, Subhodev

A2 - Parvin, Bahram

A2 - Porikli, Fatih

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Piscataway NJ USA

T2 - IEEE Winter Conference on Applications of Computer Vision 2015

Y2 - 5 January 2015 through 9 January 2015

ER -

Material classification on symmetric positive definite manifolds

Abstract

Conference

Keywords

Access to Document

Other files and links

Cite this