The University of Surrey Visual Concept Detection System at ImageCLEF@ICPR: working notes

M.A. Tahir; F. Yan; M. Barnard; M. Awais; K. Mikolajczyk; J. Kittler

doi:10.1007/978-3-642-17711-8_17

The University of Surrey Visual Concept Detection System at ImageCLEF@ICPR: working notes

M.A. Tahir
, F. Yan
, M. Barnard
, M. Awais
, K. Mikolajczyk
, J. Kittler

Research output: Contribution to conference › Paper › peer-review

Abstract

Visual concept detection is one of the most important tasks in image and video indexing. This paper describes our system in the ImageCLEF@ICPR Visual Concept Detection Task which ranked first for large-scale visual concept detection tasks in terms of Equal Error Rate (EER) and Area under Curve (AUC) and ranked third in terms of hierarchical measure. The presented approach involves state-of-the-art local descriptor computation, vector quantisation via clustering, structured scene or object representation via localised histograms of vector codes, similarity measure for kernel construction and classifier learning. The main novelty is the classifier-level and kernel-level fusion using Kernel Discriminant Analysis with RBF/Power Chi-Squared kernels obtained from various image descriptors. For 32 out of 53 individual concepts, we obtain the best performance of all 12 submissions to this task.

Original language	English
DOIs	https://doi.org/10.1007/978-3-642-17711-8_17
Publication status	Published - Aug 2010
Event	20th International Conference on Pattern Recognition (ICPR 2010) - Istanbul, Turkey Duration: 23 Aug 2010 → 26 Aug 2010

Conference

Conference	20th International Conference on Pattern Recognition (ICPR 2010)
Period	23/08/10 → 26/08/10

Bibliographical note

Note: Published in: Devrim ├£nay, Zehra ├çataltepe, Selim Aksoy (eds.) (2010) Recognizing Patterns in Signals, Speech, Images and Videos: ICPR 2010 Contests, Istanbul, Turkey, August 23-26, 2010, Contest Reports. Cham, Switzerland: Springer Nature. pp. 162-70. ISSN 0302-9743 ISBN 9783642177101

Organising Body: International Association for Pattern Recognition (IAPR).

Keywords

Computer science and informatics

Access to Document

10.1007/978-3-642-17711-8_17

https://doi.org/10.1007/978-3-642-17711-8_17

The University of Surrey Visual Concept Detection System at ImageCLEF@ICPR: working notes
Tahir, M. A., Yan, F., Barnard, M., Awais, M., Mikolajczyk, K. & Kittler, J., Aug 2010, Published in: Devrim ├£nay, Zehra ├çataltepe, Selim Aksoy (eds.) (2010) Recognizing Patterns in Signals, Speech, Images and Videos: ICPR 2010 Contests, Istanbul, Turkey, August 23-26, 2010, Contest Reports. Cham, Switzerland: Springer Nature. pp. 162-70. ISSN 0302-9743 ISBN 9783642177101 Organising Body: International Association for Pattern Recognition (IAPR). Organising Body: International Association for Pattern Recognition (IAPR)..
Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Cite this

@conference{0e20313ddd3d41dbb99b5379db77d336,

title = "The University of Surrey Visual Concept Detection System at ImageCLEF@ICPR: working notes",

abstract = "Visual concept detection is one of the most important tasks in image and video indexing. This paper describes our system in the ImageCLEF@ICPR Visual Concept Detection Task which ranked first for large-scale visual concept detection tasks in terms of Equal Error Rate (EER) and Area under Curve (AUC) and ranked third in terms of hierarchical measure. The presented approach involves state-of-the-art local descriptor computation, vector quantisation via clustering, structured scene or object representation via localised histograms of vector codes, similarity measure for kernel construction and classifier learning. The main novelty is the classifier-level and kernel-level fusion using Kernel Discriminant Analysis with RBF/Power Chi-Squared kernels obtained from various image descriptors. For 32 out of 53 individual concepts, we obtain the best performance of all 12 submissions to this task.",

keywords = "Computer science and informatics",

author = "M.A. Tahir and F. Yan and M. Barnard and M. Awais and K. Mikolajczyk and J. Kittler",

note = "Note: Published in: Devrim ├£nay, Zehra ├{\c c}ataltepe, Selim Aksoy (eds.) (2010) Recognizing Patterns in Signals, Speech, Images and Videos: ICPR 2010 Contests, Istanbul, Turkey, August 23-26, 2010, Contest Reports. Cham, Switzerland: Springer Nature. pp. 162-70. ISSN 0302-9743 ISBN 9783642177101 Organising Body: International Association for Pattern Recognition (IAPR).; 20th International Conference on Pattern Recognition (ICPR 2010) ; Conference date: 23-08-2010 Through 26-08-2010",

year = "2010",

month = aug,

doi = "10.1007/978-3-642-17711-8\_17",

language = "English",

}

TY - CONF

T1 - The University of Surrey Visual Concept Detection System at ImageCLEF@ICPR

T2 - 20th International Conference on Pattern Recognition (ICPR 2010)

AU - Tahir, M.A.

AU - Yan, F.

AU - Barnard, M.

AU - Awais, M.

AU - Mikolajczyk, K.

AU - Kittler, J.

N1 - Note: Published in: Devrim ├£nay, Zehra ├çataltepe, Selim Aksoy (eds.) (2010) Recognizing Patterns in Signals, Speech, Images and Videos: ICPR 2010 Contests, Istanbul, Turkey, August 23-26, 2010, Contest Reports. Cham, Switzerland: Springer Nature. pp. 162-70. ISSN 0302-9743 ISBN 9783642177101 Organising Body: International Association for Pattern Recognition (IAPR).

PY - 2010/8

Y1 - 2010/8

N2 - Visual concept detection is one of the most important tasks in image and video indexing. This paper describes our system in the ImageCLEF@ICPR Visual Concept Detection Task which ranked first for large-scale visual concept detection tasks in terms of Equal Error Rate (EER) and Area under Curve (AUC) and ranked third in terms of hierarchical measure. The presented approach involves state-of-the-art local descriptor computation, vector quantisation via clustering, structured scene or object representation via localised histograms of vector codes, similarity measure for kernel construction and classifier learning. The main novelty is the classifier-level and kernel-level fusion using Kernel Discriminant Analysis with RBF/Power Chi-Squared kernels obtained from various image descriptors. For 32 out of 53 individual concepts, we obtain the best performance of all 12 submissions to this task.

AB - Visual concept detection is one of the most important tasks in image and video indexing. This paper describes our system in the ImageCLEF@ICPR Visual Concept Detection Task which ranked first for large-scale visual concept detection tasks in terms of Equal Error Rate (EER) and Area under Curve (AUC) and ranked third in terms of hierarchical measure. The presented approach involves state-of-the-art local descriptor computation, vector quantisation via clustering, structured scene or object representation via localised histograms of vector codes, similarity measure for kernel construction and classifier learning. The main novelty is the classifier-level and kernel-level fusion using Kernel Discriminant Analysis with RBF/Power Chi-Squared kernels obtained from various image descriptors. For 32 out of 53 individual concepts, we obtain the best performance of all 12 submissions to this task.

KW - Computer science and informatics

U2 - 10.1007/978-3-642-17711-8_17

DO - 10.1007/978-3-642-17711-8_17

M3 - Paper

Y2 - 23 August 2010 through 26 August 2010

ER -

The University of Surrey Visual Concept Detection System at ImageCLEF@ICPR: working notes

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Fingerprint

Research output

The University of Surrey Visual Concept Detection System at ImageCLEF@ICPR: working notes

Cite this