Adaptive geometric templates for feature matching

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

Robust motion recovery in tracking multiple targets using image features is affected by difficulties in obtaining good correspondences over long sequences. Difficulties are introduced by occlusions, scale changes, as well as disappearance of features with the rotation of targets. In this work, we describe an adaptive geometric template-based method for robust motion recovery from features. A geometric template consists of nodes containing salient features (e.g., corner features). The spatial configuration of the features is modeled using a spanning tree. This paper makes the following two contributions: (i) an adaptive geometric template to model the varying number of features on a target, and (ii) an iterative data association method for the features based on the uncertainties in the estimated template structure in conjunction with its individual features. We present experimental results for tracking multiple targets over long outdoor image sequences with multiple persistent occlusions. A comparison of the results of the data association method with a standard Mahalanobis distance gating applied to individual features is also presented.

Original languageEnglish (US)
Title of host publicationProceedings 2006 IEEE International Conference on Robotics and Automation, ICRA 2006
Pages3393-3397
Number of pages5
Volume2006
DOIs
StatePublished - Dec 27 2006
Event2006 IEEE International Conference on Robotics and Automation, ICRA 2006 - Orlando, FL, United States
Duration: May 15 2006May 19 2006

Other

Other2006 IEEE International Conference on Robotics and Automation, ICRA 2006
CountryUnited States
CityOrlando, FL
Period5/15/065/19/06

Fingerprint

Recovery
Uncertainty

Keywords

  • Adaptive geometric templates, data association
  • Feature tracking
  • Kalman filtering
  • Variable dimension

Cite this

Veeraraghavan, H., Schrater, P. R., & Papanikolopoulos, N. P. (2006). Adaptive geometric templates for feature matching. In Proceedings 2006 IEEE International Conference on Robotics and Automation, ICRA 2006 (Vol. 2006, pp. 3393-3397). [1642220] https://doi.org/10.1109/ROBOT.2006.1642220

Adaptive geometric templates for feature matching. / Veeraraghavan, Harini; Schrater, Paul R; Papanikolopoulos, Nikolaos P.

Proceedings 2006 IEEE International Conference on Robotics and Automation, ICRA 2006. Vol. 2006 2006. p. 3393-3397 1642220.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Veeraraghavan, H, Schrater, PR & Papanikolopoulos, NP 2006, Adaptive geometric templates for feature matching. in Proceedings 2006 IEEE International Conference on Robotics and Automation, ICRA 2006. vol. 2006, 1642220, pp. 3393-3397, 2006 IEEE International Conference on Robotics and Automation, ICRA 2006, Orlando, FL, United States, 5/15/06. https://doi.org/10.1109/ROBOT.2006.1642220
Veeraraghavan H, Schrater PR, Papanikolopoulos NP. Adaptive geometric templates for feature matching. In Proceedings 2006 IEEE International Conference on Robotics and Automation, ICRA 2006. Vol. 2006. 2006. p. 3393-3397. 1642220 https://doi.org/10.1109/ROBOT.2006.1642220
Veeraraghavan, Harini ; Schrater, Paul R ; Papanikolopoulos, Nikolaos P. / Adaptive geometric templates for feature matching. Proceedings 2006 IEEE International Conference on Robotics and Automation, ICRA 2006. Vol. 2006 2006. pp. 3393-3397
@inproceedings{9196569d0a66456dbbf498ad810a6e1f,
title = "Adaptive geometric templates for feature matching",
abstract = "Robust motion recovery in tracking multiple targets using image features is affected by difficulties in obtaining good correspondences over long sequences. Difficulties are introduced by occlusions, scale changes, as well as disappearance of features with the rotation of targets. In this work, we describe an adaptive geometric template-based method for robust motion recovery from features. A geometric template consists of nodes containing salient features (e.g., corner features). The spatial configuration of the features is modeled using a spanning tree. This paper makes the following two contributions: (i) an adaptive geometric template to model the varying number of features on a target, and (ii) an iterative data association method for the features based on the uncertainties in the estimated template structure in conjunction with its individual features. We present experimental results for tracking multiple targets over long outdoor image sequences with multiple persistent occlusions. A comparison of the results of the data association method with a standard Mahalanobis distance gating applied to individual features is also presented.",
keywords = "Adaptive geometric templates, data association, Feature tracking, Kalman filtering, Variable dimension",
author = "Harini Veeraraghavan and Schrater, {Paul R} and Papanikolopoulos, {Nikolaos P}",
year = "2006",
month = "12",
day = "27",
doi = "10.1109/ROBOT.2006.1642220",
language = "English (US)",
isbn = "0780395069",
volume = "2006",
pages = "3393--3397",
booktitle = "Proceedings 2006 IEEE International Conference on Robotics and Automation, ICRA 2006",

}

TY - GEN

T1 - Adaptive geometric templates for feature matching

AU - Veeraraghavan, Harini

AU - Schrater, Paul R

AU - Papanikolopoulos, Nikolaos P

PY - 2006/12/27

Y1 - 2006/12/27

N2 - Robust motion recovery in tracking multiple targets using image features is affected by difficulties in obtaining good correspondences over long sequences. Difficulties are introduced by occlusions, scale changes, as well as disappearance of features with the rotation of targets. In this work, we describe an adaptive geometric template-based method for robust motion recovery from features. A geometric template consists of nodes containing salient features (e.g., corner features). The spatial configuration of the features is modeled using a spanning tree. This paper makes the following two contributions: (i) an adaptive geometric template to model the varying number of features on a target, and (ii) an iterative data association method for the features based on the uncertainties in the estimated template structure in conjunction with its individual features. We present experimental results for tracking multiple targets over long outdoor image sequences with multiple persistent occlusions. A comparison of the results of the data association method with a standard Mahalanobis distance gating applied to individual features is also presented.

AB - Robust motion recovery in tracking multiple targets using image features is affected by difficulties in obtaining good correspondences over long sequences. Difficulties are introduced by occlusions, scale changes, as well as disappearance of features with the rotation of targets. In this work, we describe an adaptive geometric template-based method for robust motion recovery from features. A geometric template consists of nodes containing salient features (e.g., corner features). The spatial configuration of the features is modeled using a spanning tree. This paper makes the following two contributions: (i) an adaptive geometric template to model the varying number of features on a target, and (ii) an iterative data association method for the features based on the uncertainties in the estimated template structure in conjunction with its individual features. We present experimental results for tracking multiple targets over long outdoor image sequences with multiple persistent occlusions. A comparison of the results of the data association method with a standard Mahalanobis distance gating applied to individual features is also presented.

KW - Adaptive geometric templates, data association

KW - Feature tracking

KW - Kalman filtering

KW - Variable dimension

UR - http://www.scopus.com/inward/record.url?scp=33845606418&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33845606418&partnerID=8YFLogxK

U2 - 10.1109/ROBOT.2006.1642220

DO - 10.1109/ROBOT.2006.1642220

M3 - Conference contribution

SN - 0780395069

SN - 9780780395060

VL - 2006

SP - 3393

EP - 3397

BT - Proceedings 2006 IEEE International Conference on Robotics and Automation, ICRA 2006

ER -