Home > Research > Publications & Outputs > Action Classification with Locality-constrained...

Links

Text available via DOI:

View graph of relations

Action Classification with Locality-constrained Linear Coding

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

Published

Standard

Action Classification with Locality-constrained Linear Coding. / Rahmani, Hossein; Mahmood, Arif; Huynh, Du Q.; Mian, Ajmal.

2014 22nd International Conference on Pattern Recognition. IEEE, 2014. p. 3511-3516.

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

Harvard

Rahmani, H, Mahmood, A, Huynh, DQ & Mian, A 2014, Action Classification with Locality-constrained Linear Coding. in 2014 22nd International Conference on Pattern Recognition. IEEE, pp. 3511-3516. https://doi.org/10.1109/ICPR.2014.604

APA

Rahmani, H., Mahmood, A., Huynh, D. Q., & Mian, A. (2014). Action Classification with Locality-constrained Linear Coding. In 2014 22nd International Conference on Pattern Recognition (pp. 3511-3516). IEEE. https://doi.org/10.1109/ICPR.2014.604

Vancouver

Rahmani H, Mahmood A, Huynh DQ, Mian A. Action Classification with Locality-constrained Linear Coding. In 2014 22nd International Conference on Pattern Recognition. IEEE. 2014. p. 3511-3516 https://doi.org/10.1109/ICPR.2014.604

Author

Rahmani, Hossein ; Mahmood, Arif ; Huynh, Du Q. ; Mian, Ajmal. / Action Classification with Locality-constrained Linear Coding. 2014 22nd International Conference on Pattern Recognition. IEEE, 2014. pp. 3511-3516

Bibtex

@inproceedings{b0d0cc16012247abaa71a9639cb61c14,
title = "Action Classification with Locality-constrained Linear Coding",
abstract = "We propose an action classification algorithm which uses Locality-constrained Linear Coding (LLC) to capture discriminative information of human body variations in each spatio-temporal subsequence of a video sequence. Our proposed method divides the input video into equally spaced overlapping spatio-temporal sub sequences, each of which is decomposed into blocks and then cells. We use the Histogram of Oriented Gradient (HOG3D) feature to encode the information in each cell. We justify the use of LLC for encoding the block descriptor by demonstrating its superiority over Sparse Coding (SC). Our sequence descriptor is obtained via a logistic regression classifier with L2 regularization. We evaluate and compare our algorithm with ten state-of-the-art algorithms on five benchmark datasets. Experimental results show that, on average, our algorithm gives better accuracy than these ten algorithms.",
author = "Hossein Rahmani and Arif Mahmood and Huynh, {Du Q.} and Ajmal Mian",
year = "2014",
doi = "10.1109/ICPR.2014.604",
language = "English",
pages = "3511--3516",
booktitle = "2014 22nd International Conference on Pattern Recognition",
publisher = "IEEE",

}

RIS

TY - GEN

T1 - Action Classification with Locality-constrained Linear Coding

AU - Rahmani, Hossein

AU - Mahmood, Arif

AU - Huynh, Du Q.

AU - Mian, Ajmal

PY - 2014

Y1 - 2014

N2 - We propose an action classification algorithm which uses Locality-constrained Linear Coding (LLC) to capture discriminative information of human body variations in each spatio-temporal subsequence of a video sequence. Our proposed method divides the input video into equally spaced overlapping spatio-temporal sub sequences, each of which is decomposed into blocks and then cells. We use the Histogram of Oriented Gradient (HOG3D) feature to encode the information in each cell. We justify the use of LLC for encoding the block descriptor by demonstrating its superiority over Sparse Coding (SC). Our sequence descriptor is obtained via a logistic regression classifier with L2 regularization. We evaluate and compare our algorithm with ten state-of-the-art algorithms on five benchmark datasets. Experimental results show that, on average, our algorithm gives better accuracy than these ten algorithms.

AB - We propose an action classification algorithm which uses Locality-constrained Linear Coding (LLC) to capture discriminative information of human body variations in each spatio-temporal subsequence of a video sequence. Our proposed method divides the input video into equally spaced overlapping spatio-temporal sub sequences, each of which is decomposed into blocks and then cells. We use the Histogram of Oriented Gradient (HOG3D) feature to encode the information in each cell. We justify the use of LLC for encoding the block descriptor by demonstrating its superiority over Sparse Coding (SC). Our sequence descriptor is obtained via a logistic regression classifier with L2 regularization. We evaluate and compare our algorithm with ten state-of-the-art algorithms on five benchmark datasets. Experimental results show that, on average, our algorithm gives better accuracy than these ten algorithms.

U2 - 10.1109/ICPR.2014.604

DO - 10.1109/ICPR.2014.604

M3 - Conference contribution/Paper

SP - 3511

EP - 3516

BT - 2014 22nd International Conference on Pattern Recognition

PB - IEEE

ER -