Object tracking and detection after occlusion via numerical hybrid local and global mode-seeking

Zhaozheng Yin, Robert Collins

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

36 Citations (Scopus)

Abstract

Given an object model and a black-box measure of similarity between the model and candidate targets, we consider visual object tracking as a numerical optimization problem. During normal tracking conditions when the object is visible from frame to frame, local optimization is used to track the local mode of the similarity measure in a parameter space of translation, rotation and scale. However, when the object becomes partially or totally occluded, such local tracking is prone to failure, especially when common prediction techniques like the Kalman filter do not provide a good estimate of object parameters in future frames. To recover from these inevitable tracking failures, we consider object detection as a global optimization problem and solve it via Adaptive Simulated Annealing (ASA), a method that avoids becoming trapped at local modes and is much faster than exhaustive search. As a Monte Carlo approach, ASA stochastically samples the parameter space, in contrast to local deterministic search. We apply cluster analysis on the sampled parameter space to redetect the object and renew the local tracker. Our numerical hybrid local and global mode-seeking tracker is validated on challenging airborne videos with heavy occlusion and large camera motions. Our approach outperforms state-of-the-art trackers on the VIVID benchmark datasets.

Original languageEnglish (US)
Title of host publication26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR
DOIs
StatePublished - Sep 23 2008
Event26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR - Anchorage, AK, United States
Duration: Jun 23 2008Jun 28 2008

Publication series

Name26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR

Other

Other26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR
CountryUnited States
CityAnchorage, AK
Period6/23/086/28/08

Fingerprint

Simulated annealing
Cluster analysis
Global optimization
Kalman filters
Cameras
Object detection

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition
  • Control and Systems Engineering

Cite this

Yin, Z., & Collins, R. (2008). Object tracking and detection after occlusion via numerical hybrid local and global mode-seeking. In 26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR [4587542] (26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR). https://doi.org/10.1109/CVPR.2008.4587542
Yin, Zhaozheng ; Collins, Robert. / Object tracking and detection after occlusion via numerical hybrid local and global mode-seeking. 26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR. 2008. (26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR).
@inproceedings{80beb47c8da14f3f85209007b0744054,
title = "Object tracking and detection after occlusion via numerical hybrid local and global mode-seeking",
abstract = "Given an object model and a black-box measure of similarity between the model and candidate targets, we consider visual object tracking as a numerical optimization problem. During normal tracking conditions when the object is visible from frame to frame, local optimization is used to track the local mode of the similarity measure in a parameter space of translation, rotation and scale. However, when the object becomes partially or totally occluded, such local tracking is prone to failure, especially when common prediction techniques like the Kalman filter do not provide a good estimate of object parameters in future frames. To recover from these inevitable tracking failures, we consider object detection as a global optimization problem and solve it via Adaptive Simulated Annealing (ASA), a method that avoids becoming trapped at local modes and is much faster than exhaustive search. As a Monte Carlo approach, ASA stochastically samples the parameter space, in contrast to local deterministic search. We apply cluster analysis on the sampled parameter space to redetect the object and renew the local tracker. Our numerical hybrid local and global mode-seeking tracker is validated on challenging airborne videos with heavy occlusion and large camera motions. Our approach outperforms state-of-the-art trackers on the VIVID benchmark datasets.",
author = "Zhaozheng Yin and Robert Collins",
year = "2008",
month = "9",
day = "23",
doi = "10.1109/CVPR.2008.4587542",
language = "English (US)",
isbn = "9781424422432",
series = "26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR",
booktitle = "26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR",

}

Yin, Z & Collins, R 2008, Object tracking and detection after occlusion via numerical hybrid local and global mode-seeking. in 26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR., 4587542, 26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR, 26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR, Anchorage, AK, United States, 6/23/08. https://doi.org/10.1109/CVPR.2008.4587542

Object tracking and detection after occlusion via numerical hybrid local and global mode-seeking. / Yin, Zhaozheng; Collins, Robert.

26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR. 2008. 4587542 (26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR).

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

TY - GEN

T1 - Object tracking and detection after occlusion via numerical hybrid local and global mode-seeking

AU - Yin, Zhaozheng

AU - Collins, Robert

PY - 2008/9/23

Y1 - 2008/9/23

N2 - Given an object model and a black-box measure of similarity between the model and candidate targets, we consider visual object tracking as a numerical optimization problem. During normal tracking conditions when the object is visible from frame to frame, local optimization is used to track the local mode of the similarity measure in a parameter space of translation, rotation and scale. However, when the object becomes partially or totally occluded, such local tracking is prone to failure, especially when common prediction techniques like the Kalman filter do not provide a good estimate of object parameters in future frames. To recover from these inevitable tracking failures, we consider object detection as a global optimization problem and solve it via Adaptive Simulated Annealing (ASA), a method that avoids becoming trapped at local modes and is much faster than exhaustive search. As a Monte Carlo approach, ASA stochastically samples the parameter space, in contrast to local deterministic search. We apply cluster analysis on the sampled parameter space to redetect the object and renew the local tracker. Our numerical hybrid local and global mode-seeking tracker is validated on challenging airborne videos with heavy occlusion and large camera motions. Our approach outperforms state-of-the-art trackers on the VIVID benchmark datasets.

AB - Given an object model and a black-box measure of similarity between the model and candidate targets, we consider visual object tracking as a numerical optimization problem. During normal tracking conditions when the object is visible from frame to frame, local optimization is used to track the local mode of the similarity measure in a parameter space of translation, rotation and scale. However, when the object becomes partially or totally occluded, such local tracking is prone to failure, especially when common prediction techniques like the Kalman filter do not provide a good estimate of object parameters in future frames. To recover from these inevitable tracking failures, we consider object detection as a global optimization problem and solve it via Adaptive Simulated Annealing (ASA), a method that avoids becoming trapped at local modes and is much faster than exhaustive search. As a Monte Carlo approach, ASA stochastically samples the parameter space, in contrast to local deterministic search. We apply cluster analysis on the sampled parameter space to redetect the object and renew the local tracker. Our numerical hybrid local and global mode-seeking tracker is validated on challenging airborne videos with heavy occlusion and large camera motions. Our approach outperforms state-of-the-art trackers on the VIVID benchmark datasets.

UR - http://www.scopus.com/inward/record.url?scp=51949091966&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=51949091966&partnerID=8YFLogxK

U2 - 10.1109/CVPR.2008.4587542

DO - 10.1109/CVPR.2008.4587542

M3 - Conference contribution

SN - 9781424422432

T3 - 26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR

BT - 26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR

ER -

Yin Z, Collins R. Object tracking and detection after occlusion via numerical hybrid local and global mode-seeking. In 26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR. 2008. 4587542. (26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR). https://doi.org/10.1109/CVPR.2008.4587542