TUTCRIS - Tampereen teknillinen yliopisto

TUTCRIS

A Parameter-free Label Propagation Algorithm for Person Identification in Stereo Videos

Tutkimustuotosvertaisarvioitu

Standard

A Parameter-free Label Propagation Algorithm for Person Identification in Stereo Videos. / Zhang, Chongsheng; Bi, Jingjun ; Liu, Changchang; Chen, Ke.

julkaisussa: Neurocomputing, Vuosikerta 218, 19.12.2016, s. 72-78.

Tutkimustuotosvertaisarvioitu

Harvard

Zhang, C, Bi, J, Liu, C & Chen, K 2016, 'A Parameter-free Label Propagation Algorithm for Person Identification in Stereo Videos', Neurocomputing, Vuosikerta. 218, Sivut 72-78. https://doi.org/10.1016/j.neucom.2016.08.069

APA

Vancouver

Author

Zhang, Chongsheng ; Bi, Jingjun ; Liu, Changchang ; Chen, Ke. / A Parameter-free Label Propagation Algorithm for Person Identification in Stereo Videos. Julkaisussa: Neurocomputing. 2016 ; Vuosikerta 218. Sivut 72-78.

Bibtex - Lataa

@article{741762b547664cb89239dc228d8bfdd9,
title = "A Parameter-free Label Propagation Algorithm for Person Identification in Stereo Videos",
abstract = "Motivated by relaxing expensive and laborious person identity annotation in stereo videos, a number of research efforts have recently been dedicated to label propagation. In this work, we propose two heuristic label propagation algorithms for annotating person identities in stereo videos under the observation that the actors in two consecutive facial images in a video are more likely to be identical. In the light of this, after adjacent video frames divided into several groups, we propose our first algorithm (i.e. ZBLC4) to automatically annotate the unlabeled images with the one having the maximum summed similarity between unlabeled and labeled images in each group in the parameter-free manner. Moreover, to cope with singleton groups, an additional classifier is introduced into ZBLC4 algorithm to mitigate the suffering of unreliable prediction dependent on neighbors. We conduct experiments on three publicly-benchmarking stereo videos, demonstrating that our algorithms are superior to the state-of-the-arts.",
author = "Chongsheng Zhang and Jingjun Bi and Changchang Liu and Ke Chen",
year = "2016",
month = "12",
day = "19",
doi = "10.1016/j.neucom.2016.08.069",
language = "English",
volume = "218",
pages = "72--78",
journal = "Neurocomputing",
issn = "0925-2312",
publisher = "Elsevier Science B.V.",

}

RIS (suitable for import to EndNote) - Lataa

TY - JOUR

T1 - A Parameter-free Label Propagation Algorithm for Person Identification in Stereo Videos

AU - Zhang, Chongsheng

AU - Bi, Jingjun

AU - Liu, Changchang

AU - Chen, Ke

PY - 2016/12/19

Y1 - 2016/12/19

N2 - Motivated by relaxing expensive and laborious person identity annotation in stereo videos, a number of research efforts have recently been dedicated to label propagation. In this work, we propose two heuristic label propagation algorithms for annotating person identities in stereo videos under the observation that the actors in two consecutive facial images in a video are more likely to be identical. In the light of this, after adjacent video frames divided into several groups, we propose our first algorithm (i.e. ZBLC4) to automatically annotate the unlabeled images with the one having the maximum summed similarity between unlabeled and labeled images in each group in the parameter-free manner. Moreover, to cope with singleton groups, an additional classifier is introduced into ZBLC4 algorithm to mitigate the suffering of unreliable prediction dependent on neighbors. We conduct experiments on three publicly-benchmarking stereo videos, demonstrating that our algorithms are superior to the state-of-the-arts.

AB - Motivated by relaxing expensive and laborious person identity annotation in stereo videos, a number of research efforts have recently been dedicated to label propagation. In this work, we propose two heuristic label propagation algorithms for annotating person identities in stereo videos under the observation that the actors in two consecutive facial images in a video are more likely to be identical. In the light of this, after adjacent video frames divided into several groups, we propose our first algorithm (i.e. ZBLC4) to automatically annotate the unlabeled images with the one having the maximum summed similarity between unlabeled and labeled images in each group in the parameter-free manner. Moreover, to cope with singleton groups, an additional classifier is introduced into ZBLC4 algorithm to mitigate the suffering of unreliable prediction dependent on neighbors. We conduct experiments on three publicly-benchmarking stereo videos, demonstrating that our algorithms are superior to the state-of-the-arts.

U2 - 10.1016/j.neucom.2016.08.069

DO - 10.1016/j.neucom.2016.08.069

M3 - Article

VL - 218

SP - 72

EP - 78

JO - Neurocomputing

JF - Neurocomputing

SN - 0925-2312

ER -