Accelerometer localization in the view of a stationary camera

S. Stein, Stephen McKenna

Research output: Chapter in Book/Report/Conference proceedingChapter

4 Citations (Scopus)

Abstract

This paper addresses the problem of localizing an accelerometer in the view of a stationary camera as a first step towards multi-model activity recognition. This problem is challenging as accelerometers are visually occluded, they measure proper acceleration including effects of gravity and their orientation is unknown and changes over time relative to camera viewpoint. Accelerometers are localized by matching acceleration estimated along visual point trajectories to accelerometer data. Trajectories are constructed from point feature tracking (KLT) and by grid sampling from a dense flow field. We also construct 3D trajectories with visual depth information. The similarity between accelerometer data and a trajectory is computed by counting the number of frames in which the norms of accelerations in both sequences exceed a threshold. For quantitative evaluation we collected a challenging dataset consisting of video and accelerometer data of a person preparing a mixed salad with accelerometer-equipped kitchen utensils. Trajectories from dense optical flow yielded a higher localization accuracy compared to point feature tracking.
Original languageEnglish
Title of host publicationProceedings of the 2012 9th Conference on Computer and Robot Vision, CRV 2012
Pages109-116
Number of pages8
DOIs
Publication statusPublished - 1 Jan 2012

Fingerprint

Accelerometers
Cameras
Trajectories
Kitchens
Optical flows
Flow fields
Gravitation
Sampling

Cite this

Stein, S., & McKenna, S. (2012). Accelerometer localization in the view of a stationary camera. In Proceedings of the 2012 9th Conference on Computer and Robot Vision, CRV 2012 (pp. 109-116) https://doi.org/10.1109/CRV.2012.22
Stein, S. ; McKenna, Stephen. / Accelerometer localization in the view of a stationary camera. Proceedings of the 2012 9th Conference on Computer and Robot Vision, CRV 2012. 2012. pp. 109-116
@inbook{4587fa74dd59402e931d10a4536410bf,
title = "Accelerometer localization in the view of a stationary camera",
abstract = "This paper addresses the problem of localizing an accelerometer in the view of a stationary camera as a first step towards multi-model activity recognition. This problem is challenging as accelerometers are visually occluded, they measure proper acceleration including effects of gravity and their orientation is unknown and changes over time relative to camera viewpoint. Accelerometers are localized by matching acceleration estimated along visual point trajectories to accelerometer data. Trajectories are constructed from point feature tracking (KLT) and by grid sampling from a dense flow field. We also construct 3D trajectories with visual depth information. The similarity between accelerometer data and a trajectory is computed by counting the number of frames in which the norms of accelerations in both sequences exceed a threshold. For quantitative evaluation we collected a challenging dataset consisting of video and accelerometer data of a person preparing a mixed salad with accelerometer-equipped kitchen utensils. Trajectories from dense optical flow yielded a higher localization accuracy compared to point feature tracking.",
author = "S. Stein and Stephen McKenna",
year = "2012",
month = "1",
day = "1",
doi = "10.1109/CRV.2012.22",
language = "English",
isbn = "9780769546834",
pages = "109--116",
booktitle = "Proceedings of the 2012 9th Conference on Computer and Robot Vision, CRV 2012",

}

Stein, S & McKenna, S 2012, Accelerometer localization in the view of a stationary camera. in Proceedings of the 2012 9th Conference on Computer and Robot Vision, CRV 2012. pp. 109-116. https://doi.org/10.1109/CRV.2012.22

Accelerometer localization in the view of a stationary camera. / Stein, S.; McKenna, Stephen.

Proceedings of the 2012 9th Conference on Computer and Robot Vision, CRV 2012. 2012. p. 109-116.

Research output: Chapter in Book/Report/Conference proceedingChapter

TY - CHAP

T1 - Accelerometer localization in the view of a stationary camera

AU - Stein, S.

AU - McKenna, Stephen

PY - 2012/1/1

Y1 - 2012/1/1

N2 - This paper addresses the problem of localizing an accelerometer in the view of a stationary camera as a first step towards multi-model activity recognition. This problem is challenging as accelerometers are visually occluded, they measure proper acceleration including effects of gravity and their orientation is unknown and changes over time relative to camera viewpoint. Accelerometers are localized by matching acceleration estimated along visual point trajectories to accelerometer data. Trajectories are constructed from point feature tracking (KLT) and by grid sampling from a dense flow field. We also construct 3D trajectories with visual depth information. The similarity between accelerometer data and a trajectory is computed by counting the number of frames in which the norms of accelerations in both sequences exceed a threshold. For quantitative evaluation we collected a challenging dataset consisting of video and accelerometer data of a person preparing a mixed salad with accelerometer-equipped kitchen utensils. Trajectories from dense optical flow yielded a higher localization accuracy compared to point feature tracking.

AB - This paper addresses the problem of localizing an accelerometer in the view of a stationary camera as a first step towards multi-model activity recognition. This problem is challenging as accelerometers are visually occluded, they measure proper acceleration including effects of gravity and their orientation is unknown and changes over time relative to camera viewpoint. Accelerometers are localized by matching acceleration estimated along visual point trajectories to accelerometer data. Trajectories are constructed from point feature tracking (KLT) and by grid sampling from a dense flow field. We also construct 3D trajectories with visual depth information. The similarity between accelerometer data and a trajectory is computed by counting the number of frames in which the norms of accelerations in both sequences exceed a threshold. For quantitative evaluation we collected a challenging dataset consisting of video and accelerometer data of a person preparing a mixed salad with accelerometer-equipped kitchen utensils. Trajectories from dense optical flow yielded a higher localization accuracy compared to point feature tracking.

UR - http://www.scopus.com/inward/record.url?scp=84878348844&partnerID=8YFLogxK

U2 - 10.1109/CRV.2012.22

DO - 10.1109/CRV.2012.22

M3 - Chapter

SN - 9780769546834

SP - 109

EP - 116

BT - Proceedings of the 2012 9th Conference on Computer and Robot Vision, CRV 2012

ER -

Stein S, McKenna S. Accelerometer localization in the view of a stationary camera. In Proceedings of the 2012 9th Conference on Computer and Robot Vision, CRV 2012. 2012. p. 109-116 https://doi.org/10.1109/CRV.2012.22