Home > Research > Publications & Outputs > Global Regularizer and Temporal-aware Cross-ent...

Associated organisational unit

Electronic data

  • 0050

    Accepted author manuscript, 1.06 MB, PDF document

Links

View graph of relations

Global Regularizer and Temporal-aware Cross-entropy for Skeleton-based Early Action Recognition

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Published

Standard

Global Regularizer and Temporal-aware Cross-entropy for Skeleton-based Early Action Recognition. / Ke, Qiuhong; Liu, Jun; Bennamoun, Mohammed et al.
Computer Vision - ACCV 2018: 14th Asian Conference on Computer Vision. Vol. IV Springer, 2018.

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Harvard

Ke, Q, Liu, J, Bennamoun, M, Rahmani, H, An, S, Sohel, F & Boussaid, F 2018, Global Regularizer and Temporal-aware Cross-entropy for Skeleton-based Early Action Recognition. in Computer Vision - ACCV 2018: 14th Asian Conference on Computer Vision. vol. IV, Springer, 14th Asian Conference on Computer Vision (ACCV), Perth, Western Australia, Australia, 4/12/18. <https://www.springerprofessional.de/en/global-regularizer-and-temporal-aware-cross-entropy-for-skeleton/16749394>

APA

Ke, Q., Liu, J., Bennamoun, M., Rahmani, H., An, S., Sohel, F., & Boussaid, F. (2018). Global Regularizer and Temporal-aware Cross-entropy for Skeleton-based Early Action Recognition. In Computer Vision - ACCV 2018: 14th Asian Conference on Computer Vision (Vol. IV). Springer. https://www.springerprofessional.de/en/global-regularizer-and-temporal-aware-cross-entropy-for-skeleton/16749394

Vancouver

Ke Q, Liu J, Bennamoun M, Rahmani H, An S, Sohel F et al. Global Regularizer and Temporal-aware Cross-entropy for Skeleton-based Early Action Recognition. In Computer Vision - ACCV 2018: 14th Asian Conference on Computer Vision. Vol. IV. Springer. 2018

Author

Ke, Qiuhong ; Liu, Jun ; Bennamoun, Mohammed et al. / Global Regularizer and Temporal-aware Cross-entropy for Skeleton-based Early Action Recognition. Computer Vision - ACCV 2018: 14th Asian Conference on Computer Vision. Vol. IV Springer, 2018.

Bibtex

@inproceedings{158303a69b434ab4ada7150d4cb4bc89,
title = "Global Regularizer and Temporal-aware Cross-entropy for Skeleton-based Early Action Recognition",
abstract = "In this paper, we propose a new approach to recognize the class label of an action before this action is fully performed based on skeleton sequences. Compared to action recognition which uses fully observed action sequences, early action recognition with partial sequences is much more challenging mainly due to: (1) the global information of a long-term action is not available in the partial sequence, and (2) the partial sequences at different observation ratios of an action contain a number of sub-actions with diverse motion information. To address the first challenge, we introduce a global regularizer to learn a hidden feature space, where the statistical properties of the partial sequences are similar to those of the full sequences. We introduce a temporal-aware cross-entropy to address the second challenge and achieve better prediction performance. We evaluate the proposed method on three challenging skeleton datasets. Experimental results show the superiority of the proposed method for skeleton-based early action recognition.",
author = "Qiuhong Ke and Jun Liu and Mohammed Bennamoun and Hossein Rahmani and Senjian An and Ferdous Sohel and Farid Boussaid",
year = "2018",
month = dec,
day = "2",
language = "English",
isbn = "9783030208691",
volume = "IV",
booktitle = "Computer Vision - ACCV 2018",
publisher = "Springer",
note = "14th Asian Conference on Computer Vision (ACCV) ; Conference date: 04-12-2018",

}

RIS

TY - GEN

T1 - Global Regularizer and Temporal-aware Cross-entropy for Skeleton-based Early Action Recognition

AU - Ke, Qiuhong

AU - Liu, Jun

AU - Bennamoun, Mohammed

AU - Rahmani, Hossein

AU - An, Senjian

AU - Sohel, Ferdous

AU - Boussaid, Farid

PY - 2018/12/2

Y1 - 2018/12/2

N2 - In this paper, we propose a new approach to recognize the class label of an action before this action is fully performed based on skeleton sequences. Compared to action recognition which uses fully observed action sequences, early action recognition with partial sequences is much more challenging mainly due to: (1) the global information of a long-term action is not available in the partial sequence, and (2) the partial sequences at different observation ratios of an action contain a number of sub-actions with diverse motion information. To address the first challenge, we introduce a global regularizer to learn a hidden feature space, where the statistical properties of the partial sequences are similar to those of the full sequences. We introduce a temporal-aware cross-entropy to address the second challenge and achieve better prediction performance. We evaluate the proposed method on three challenging skeleton datasets. Experimental results show the superiority of the proposed method for skeleton-based early action recognition.

AB - In this paper, we propose a new approach to recognize the class label of an action before this action is fully performed based on skeleton sequences. Compared to action recognition which uses fully observed action sequences, early action recognition with partial sequences is much more challenging mainly due to: (1) the global information of a long-term action is not available in the partial sequence, and (2) the partial sequences at different observation ratios of an action contain a number of sub-actions with diverse motion information. To address the first challenge, we introduce a global regularizer to learn a hidden feature space, where the statistical properties of the partial sequences are similar to those of the full sequences. We introduce a temporal-aware cross-entropy to address the second challenge and achieve better prediction performance. We evaluate the proposed method on three challenging skeleton datasets. Experimental results show the superiority of the proposed method for skeleton-based early action recognition.

M3 - Conference contribution/Paper

SN - 9783030208691

VL - IV

BT - Computer Vision - ACCV 2018

PB - Springer

T2 - 14th Asian Conference on Computer Vision (ACCV)

Y2 - 4 December 2018

ER -