Stabilizing Motion Tracking Using Retrieved Motion Priors

verfasst von: Andreas Baak, Bodo Rosenhahn, Meinard Müller, Hans Peter Seidel
Abstract: In this paper, we introduce a novel iterative motion tracking framework that combines 3D tracking techniques with motion retrieval for stabilizing markerless human motion capturing. The basic idea is to start human tracking without prior knowledge about the performed actions. The resulting 3D motion sequences, which may be corrupted due to tracking errors, are locally classified according to available motion categories. Depending on the classification result, a retrieval system supplies suitable motion priors, which are then used to regularize and stabilize the tracking in the next iteration step. Experiments with the HumanEVA-II benchmark show that tracking and classification are remarkably improved after few iterations.
Organisationseinheit(en): Institut für Informationsverarbeitung
Externe Organisation(en): Universität des Saarlandes
Typ: Konferenzaufsatz in Fachzeitschrift
Journal: Proceedings of the IEEE International Conference on Computer Vision
Seiten: 1428-1435
Anzahl der Seiten: 8
ISSN: 1550-5499
Publikationsdatum: 2009
Publikationsstatus: Veröffentlicht
Peer-reviewed: Ja
ASJC Scopus Sachgebiete: Software, Maschinelles Sehen und Mustererkennung
Elektronische Version(en): https://doi.org/10.1109/ICCV.2009.5459291 (Zugang: Geschlossen)

BibTeX

@article{ac4a88fd555f4bdf80ca1be1656c0229,
title = "Stabilizing Motion Tracking Using Retrieved Motion Priors",
abstract = "In this paper, we introduce a novel iterative motion tracking framework that combines 3D tracking techniques with motion retrieval for stabilizing markerless human motion capturing. The basic idea is to start human tracking without prior knowledge about the performed actions. The resulting 3D motion sequences, which may be corrupted due to tracking errors, are locally classified according to available motion categories. Depending on the classification result, a retrieval system supplies suitable motion priors, which are then used to regularize and stabilize the tracking in the next iteration step. Experiments with the HumanEVA-II benchmark show that tracking and classification are remarkably improved after few iterations.",
author = "Andreas Baak and Bodo Rosenhahn and Meinard M{\"u}ller and Seidel, {Hans Peter}",
note = "Funding Information: Acknowledgments. The first and second author are funded by the German Research Foundation (DFG CL 64/5-1, RO 2497/6-1), the third author by the Cluster of Excellence on Multimodal Computing and Interaction. ; 12th International Conference on Computer Vision, ICCV 2009 ; Conference date: 29-09-2009 Through 02-10-2009",
year = "2009",
doi = "10.1109/ICCV.2009.5459291",
language = "English",
pages = "1428--1435",
journal = "Proceedings of the IEEE International Conference on Computer Vision",
issn = "1550-5499",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
}