A Novel 3D Gradient LBP Descriptor for Action Recognition

Zhaoyang GUO; Xin'an WANG; Bo WANG; Zheng XIE

doi:10.1587/transinf.2017EDL8006

A Novel 3D Gradient LBP Descriptor for Action Recognition

Zhaoyang GUO, Xin'an WANG, Bo WANG, Zheng XIE

Full Text Views

0

Cite this

Summary :

In the field of action recognition, Spatio-Temporal Interest Points (STIPs)-based features have shown high efficiency and robustness. However, most of state-of-the-art work to describe STIPs, they typically focus on 2-dimensions (2D) images, which ignore information in 3D spatio-temporal space. Besides, the compact representation of descriptors should be considered due to the costs of storage and computational time. In this paper, a novel local descriptor named 3D Gradient LBP is proposed, which extends the traditional descriptor Local Binary Patterns (LBP) into 3D spatio-temporal space. The proposed descriptor takes advantage of the neighbourhood information of cuboids in three dimensions, which accounts for its excellent descriptive power for the distribution of grey-level space. Experiments on three challenging datasets (KTH, Weizmann and UT Interaction) validate the effectiveness of our approach in the recognition of human actions.

Publication: IEICE TRANSACTIONS on Information Vol.E100-D No.6 pp.1388-1392

Publication Date: 2017/06/01

Publicized: 2017/03/02

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2017EDL8006

Type of Manuscript: LETTER

Category: Image Recognition, Computer Vision

Authors

Zhaoyang GUO
  Shenzhen Graduate School, Peking University
Xin'an WANG
  Shenzhen Graduate School, Peking University
Bo WANG
  Shenzhen Graduate School, Peking University
Zheng XIE
  Shenzhen Graduate School, Peking University

Keyword

action recognition, spatio-temporal interest points, local binary pattern

Cite this

Copy

Zhaoyang GUO, Xin'an WANG, Bo WANG, Zheng XIE, "A Novel 3D Gradient LBP Descriptor for Action Recognition" in IEICE TRANSACTIONS on Information, vol. E100-D, no. 6, pp. 1388-1392, June 2017, doi: 10.1587/transinf.2017EDL8006.
Abstract: In the field of action recognition, Spatio-Temporal Interest Points (STIPs)-based features have shown high efficiency and robustness. However, most of state-of-the-art work to describe STIPs, they typically focus on 2-dimensions (2D) images, which ignore information in 3D spatio-temporal space. Besides, the compact representation of descriptors should be considered due to the costs of storage and computational time. In this paper, a novel local descriptor named 3D Gradient LBP is proposed, which extends the traditional descriptor Local Binary Patterns (LBP) into 3D spatio-temporal space. The proposed descriptor takes advantage of the neighbourhood information of cuboids in three dimensions, which accounts for its excellent descriptive power for the distribution of grey-level space. Experiments on three challenging datasets (KTH, Weizmann and UT Interaction) validate the effectiveness of our approach in the recognition of human actions.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2017EDL8006/_p

Copy

@ARTICLE{e100-d_6_1388,
author={Zhaoyang GUO, Xin'an WANG, Bo WANG, Zheng XIE, },
journal={IEICE TRANSACTIONS on Information},
title={A Novel 3D Gradient LBP Descriptor for Action Recognition},
year={2017},
volume={E100-D},
number={6},
pages={1388-1392},
abstract={In the field of action recognition, Spatio-Temporal Interest Points (STIPs)-based features have shown high efficiency and robustness. However, most of state-of-the-art work to describe STIPs, they typically focus on 2-dimensions (2D) images, which ignore information in 3D spatio-temporal space. Besides, the compact representation of descriptors should be considered due to the costs of storage and computational time. In this paper, a novel local descriptor named 3D Gradient LBP is proposed, which extends the traditional descriptor Local Binary Patterns (LBP) into 3D spatio-temporal space. The proposed descriptor takes advantage of the neighbourhood information of cuboids in three dimensions, which accounts for its excellent descriptive power for the distribution of grey-level space. Experiments on three challenging datasets (KTH, Weizmann and UT Interaction) validate the effectiveness of our approach in the recognition of human actions.},
keywords={},
doi={10.1587/transinf.2017EDL8006},
ISSN={1745-1361},
month={June},}

Copy

TY - JOUR
TI - A Novel 3D Gradient LBP Descriptor for Action Recognition
T2 - IEICE TRANSACTIONS on Information
SP - 1388
EP - 1392
AU - Zhaoyang GUO
AU - Xin'an WANG
AU - Bo WANG
AU - Zheng XIE
PY - 2017
DO - 10.1587/transinf.2017EDL8006
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E100-D
IS - 6
JA - IEICE TRANSACTIONS on Information
Y1 - June 2017
AB - In the field of action recognition, Spatio-Temporal Interest Points (STIPs)-based features have shown high efficiency and robustness. However, most of state-of-the-art work to describe STIPs, they typically focus on 2-dimensions (2D) images, which ignore information in 3D spatio-temporal space. Besides, the compact representation of descriptors should be considered due to the costs of storage and computational time. In this paper, a novel local descriptor named 3D Gradient LBP is proposed, which extends the traditional descriptor Local Binary Patterns (LBP) into 3D spatio-temporal space. The proposed descriptor takes advantage of the neighbourhood information of cuboids in three dimensions, which accounts for its excellent descriptive power for the distribution of grey-level space. Experiments on three challenging datasets (KTH, Weizmann and UT Interaction) validate the effectiveness of our approach in the recognition of human actions.
ER -