Margin based likelihood map fusion for target tracking

Jing Peng, Guna Seetharaman

Research output: Contribution to conferencePaperResearchpeer-review

1 Citation (Scopus)

Abstract

Visual object recognition and tracking can be formulated as an object-background classification problem. Since combining multi-modal information is known to exponentially quicken classification, often different features are used to create a set of representations for a pixel or target object. Each of the representations generates a probability of that pixel being part of the target object or scene background. Thus, how to combine these views to effectively exploit multi-modal information for classification becomes a key issue. We propose a margin based fusion technique for exploiting these heterogeneous features for classification, thus tracking. All representations contribute to classification on their learned con# dence scores (weights). As a result of optimally combining multi-modal information or evidence, discriminant object and background information is preserved, while ambiguous information is discarded. We provide experimental results that show its performance against competing techniques.

Original languageEnglish
Pages2292-2295
Number of pages4
DOIs
StatePublished - 1 Dec 2012
Event2012 32nd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2012 - Munich, Germany
Duration: 22 Jul 201227 Jul 2012

Other

Other2012 32nd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2012
CountryGermany
CityMunich
Period22/07/1227/07/12

Fingerprint

Target tracking
Fusion reactions
pixel
Pixels
Object recognition

Keywords

  • Classification
  • Fusion
  • Large margin

Cite this

Peng, J., & Seetharaman, G. (2012). Margin based likelihood map fusion for target tracking. 2292-2295. Paper presented at 2012 32nd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2012, Munich, Germany. https://doi.org/10.1109/IGARSS.2012.6351037
Peng, Jing ; Seetharaman, Guna. / Margin based likelihood map fusion for target tracking. Paper presented at 2012 32nd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2012, Munich, Germany.4 p.
@conference{08977a69507c4db38f5ad6753a076c7f,
title = "Margin based likelihood map fusion for target tracking",
abstract = "Visual object recognition and tracking can be formulated as an object-background classification problem. Since combining multi-modal information is known to exponentially quicken classification, often different features are used to create a set of representations for a pixel or target object. Each of the representations generates a probability of that pixel being part of the target object or scene background. Thus, how to combine these views to effectively exploit multi-modal information for classification becomes a key issue. We propose a margin based fusion technique for exploiting these heterogeneous features for classification, thus tracking. All representations contribute to classification on their learned con# dence scores (weights). As a result of optimally combining multi-modal information or evidence, discriminant object and background information is preserved, while ambiguous information is discarded. We provide experimental results that show its performance against competing techniques.",
keywords = "Classification, Fusion, Large margin",
author = "Jing Peng and Guna Seetharaman",
year = "2012",
month = "12",
day = "1",
doi = "10.1109/IGARSS.2012.6351037",
language = "English",
pages = "2292--2295",
note = "null ; Conference date: 22-07-2012 Through 27-07-2012",

}

Peng, J & Seetharaman, G 2012, 'Margin based likelihood map fusion for target tracking' Paper presented at 2012 32nd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2012, Munich, Germany, 22/07/12 - 27/07/12, pp. 2292-2295. https://doi.org/10.1109/IGARSS.2012.6351037

Margin based likelihood map fusion for target tracking. / Peng, Jing; Seetharaman, Guna.

2012. 2292-2295 Paper presented at 2012 32nd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2012, Munich, Germany.

Research output: Contribution to conferencePaperResearchpeer-review

TY - CONF

T1 - Margin based likelihood map fusion for target tracking

AU - Peng, Jing

AU - Seetharaman, Guna

PY - 2012/12/1

Y1 - 2012/12/1

N2 - Visual object recognition and tracking can be formulated as an object-background classification problem. Since combining multi-modal information is known to exponentially quicken classification, often different features are used to create a set of representations for a pixel or target object. Each of the representations generates a probability of that pixel being part of the target object or scene background. Thus, how to combine these views to effectively exploit multi-modal information for classification becomes a key issue. We propose a margin based fusion technique for exploiting these heterogeneous features for classification, thus tracking. All representations contribute to classification on their learned con# dence scores (weights). As a result of optimally combining multi-modal information or evidence, discriminant object and background information is preserved, while ambiguous information is discarded. We provide experimental results that show its performance against competing techniques.

AB - Visual object recognition and tracking can be formulated as an object-background classification problem. Since combining multi-modal information is known to exponentially quicken classification, often different features are used to create a set of representations for a pixel or target object. Each of the representations generates a probability of that pixel being part of the target object or scene background. Thus, how to combine these views to effectively exploit multi-modal information for classification becomes a key issue. We propose a margin based fusion technique for exploiting these heterogeneous features for classification, thus tracking. All representations contribute to classification on their learned con# dence scores (weights). As a result of optimally combining multi-modal information or evidence, discriminant object and background information is preserved, while ambiguous information is discarded. We provide experimental results that show its performance against competing techniques.

KW - Classification

KW - Fusion

KW - Large margin

UR - http://www.scopus.com/inward/record.url?scp=84873183624&partnerID=8YFLogxK

U2 - 10.1109/IGARSS.2012.6351037

DO - 10.1109/IGARSS.2012.6351037

M3 - Paper

SP - 2292

EP - 2295

ER -

Peng J, Seetharaman G. Margin based likelihood map fusion for target tracking. 2012. Paper presented at 2012 32nd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2012, Munich, Germany. https://doi.org/10.1109/IGARSS.2012.6351037