Kronecker decomposition for image classification

Sabrina Fontanella; Antonio J. Rodríguez-Sánchez; Justus Piater; Sandor Szedmak

doi:10.1007/978-3-319-44564-9_11

Kronecker decomposition for image classification

Sabrina Fontanella^*
, Antonio J. Rodríguez-Sánchez
, Justus Piater
, Sandor Szedmak

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

2 Citations (Scopus)

Abstract

We propose an image decomposition technique that captures the structure of a scene. An image is decomposed into a matrix that represents the adjacency between the elements of the image and their distance. Images decomposed this way are then classified using a maximum margin regression (MMR) approach where the normal vector of the separating hyperplane maps the input feature vectors into the outputs vectors. Multiclass and multilabel classification are native to MMR, unlike other more classical maximum margin approaches, like SVM. We have tested our approach with the ImageCLEF 2015 multi-label classification task, Pascal VOC and Flickr dataset.

Original language	English
Title of host publication	Experimental IR Meets Multilinguality, Multimodality, and Interaction - 7th International Conference of the CLEF Association, CLEF 2016, Proceedings
Publisher	Springer
Pages	137-149
Number of pages	13
Volume	9822
ISBN (Print)	9783319445632
DOIs	https://doi.org/10.1007/978-3-319-44564-9_11
Publication status	Published - 2016
MoE publication type	A4 Conference publication
Event	Conference and Labs of the Evaluation Forum - Evora, Portugal Duration: 5 Sept 2016 → 8 Sept 2016 Conference number: 7

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	9822
ISSN (Print)	03029743
ISSN (Electronic)	16113349

Conference

Conference	Conference and Labs of the Evaluation Forum
Abbreviated title	CLEF
Country/Territory	Portugal
City	Evora
Period	05/09/2016 → 08/09/2016

Keywords

ImageCLEF
Kronecker decomposition
Maximum margin
Medical images
MMR
Multi-label classification
SVM

Access to Document

10.1007/978-3-319-44564-9_11

Cite this

Fontanella, S., Rodríguez-Sánchez, A. J., Piater, J., & Szedmak, S. (2016). Kronecker decomposition for image classification. In Experimental IR Meets Multilinguality, Multimodality, and Interaction - 7th International Conference of the CLEF Association, CLEF 2016, Proceedings (Vol. 9822, pp. 137-149). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9822). Springer. https://doi.org/10.1007/978-3-319-44564-9_11

Fontanella, Sabrina ; Rodríguez-Sánchez, Antonio J. ; Piater, Justus et al. / Kronecker decomposition for image classification. Experimental IR Meets Multilinguality, Multimodality, and Interaction - 7th International Conference of the CLEF Association, CLEF 2016, Proceedings. Vol. 9822 Springer, 2016. pp. 137-149 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{bfde039345c5404a8433a1fa1eca5cde,

title = "Kronecker decomposition for image classification",

abstract = "We propose an image decomposition technique that captures the structure of a scene. An image is decomposed into a matrix that represents the adjacency between the elements of the image and their distance. Images decomposed this way are then classified using a maximum margin regression (MMR) approach where the normal vector of the separating hyperplane maps the input feature vectors into the outputs vectors. Multiclass and multilabel classification are native to MMR, unlike other more classical maximum margin approaches, like SVM. We have tested our approach with the ImageCLEF 2015 multi-label classification task, Pascal VOC and Flickr dataset.",

keywords = "ImageCLEF, Kronecker decomposition, Maximum margin, Medical images, MMR, Multi-label classification, SVM",

author = "Sabrina Fontanella and Rodr{\'i}guez-S{\'a}nchez, \{Antonio J.\} and Justus Piater and Sandor Szedmak",

year = "2016",

doi = "10.1007/978-3-319-44564-9\_11",

language = "English",

isbn = "9783319445632",

volume = "9822",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer",

pages = "137--149",

booktitle = "Experimental IR Meets Multilinguality, Multimodality, and Interaction - 7th International Conference of the CLEF Association, CLEF 2016, Proceedings",

address = "Germany",

note = "Conference and Labs of the Evaluation Forum, CLEF ; Conference date: 05-09-2016 Through 08-09-2016",

}

Fontanella, S, Rodríguez-Sánchez, AJ, Piater, J & Szedmak, S 2016, Kronecker decomposition for image classification. in Experimental IR Meets Multilinguality, Multimodality, and Interaction - 7th International Conference of the CLEF Association, CLEF 2016, Proceedings. vol. 9822, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9822, Springer, pp. 137-149, Conference and Labs of the Evaluation Forum, Evora, Portugal, 05/09/2016. https://doi.org/10.1007/978-3-319-44564-9_11

Kronecker decomposition for image classification. / Fontanella, Sabrina; Rodríguez-Sánchez, Antonio J.; Piater, Justus et al.
Experimental IR Meets Multilinguality, Multimodality, and Interaction - 7th International Conference of the CLEF Association, CLEF 2016, Proceedings. Vol. 9822 Springer, 2016. p. 137-149 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9822).

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - Kronecker decomposition for image classification

AU - Fontanella, Sabrina

AU - Rodríguez-Sánchez, Antonio J.

AU - Piater, Justus

AU - Szedmak, Sandor

N1 - Conference code: 7

PY - 2016

Y1 - 2016

N2 - We propose an image decomposition technique that captures the structure of a scene. An image is decomposed into a matrix that represents the adjacency between the elements of the image and their distance. Images decomposed this way are then classified using a maximum margin regression (MMR) approach where the normal vector of the separating hyperplane maps the input feature vectors into the outputs vectors. Multiclass and multilabel classification are native to MMR, unlike other more classical maximum margin approaches, like SVM. We have tested our approach with the ImageCLEF 2015 multi-label classification task, Pascal VOC and Flickr dataset.

AB - We propose an image decomposition technique that captures the structure of a scene. An image is decomposed into a matrix that represents the adjacency between the elements of the image and their distance. Images decomposed this way are then classified using a maximum margin regression (MMR) approach where the normal vector of the separating hyperplane maps the input feature vectors into the outputs vectors. Multiclass and multilabel classification are native to MMR, unlike other more classical maximum margin approaches, like SVM. We have tested our approach with the ImageCLEF 2015 multi-label classification task, Pascal VOC and Flickr dataset.

KW - ImageCLEF

KW - Kronecker decomposition

KW - Maximum margin

KW - Medical images

KW - MMR

KW - Multi-label classification

KW - SVM

UR - https://www.scopus.com/pages/publications/84984817864

U2 - 10.1007/978-3-319-44564-9_11

DO - 10.1007/978-3-319-44564-9_11

M3 - Conference article in proceedings

AN - SCOPUS:84984817864

SN - 9783319445632

VL - 9822

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 137

EP - 149

BT - Experimental IR Meets Multilinguality, Multimodality, and Interaction - 7th International Conference of the CLEF Association, CLEF 2016, Proceedings

PB - Springer

T2 - Conference and Labs of the Evaluation Forum

Y2 - 5 September 2016 through 8 September 2016

ER -

Fontanella S, Rodríguez-Sánchez AJ, Piater J, Szedmak S. Kronecker decomposition for image classification. In Experimental IR Meets Multilinguality, Multimodality, and Interaction - 7th International Conference of the CLEF Association, CLEF 2016, Proceedings. Vol. 9822. Springer. 2016. p. 137-149. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-44564-9_11

Kronecker decomposition for image classification

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this