Action Recognition by Improved Dense Trajectories

Authors

Peizhen Xu, School of Electrical Engineering, Southwest Jiaotong University, Chengdu 610031, China;
Zhibin Yu, School of Electrical Engineering, Southwest Jiaotong University, Chengdu 610031, China;
Weidong Jin, School of Electrical Engineering, Southwest Jiaotong University, Chengdu 610031, China;
Haiying Jiang, School of Electrical Engineering, Southwest Jiaotong University, Chengdu 610031, China;

Abstract

Abstract: In recent years, with strong robustness to fast irregular motion, the method of human motion representation based on dense trajectories has been used more and more in the field of behavior recognition. However, the relative motion of the background caused by the motion of the camera has a great influence on the extraction of the trajectories. In order to estimate the camera motion, the speed up robust feature (SURF) descriptor was used to match the feature points of each frame. Since the human motion and the camera motion were not same, human detection was added to remove inconsistent matches. Finally, multi instance learning (MIL) was used to classify and recognize the behavior. Experiment results demonstrate the effectiveness of the approach on the UT-interaction dataset.

Recommended Citation

Xu, Peizhen; Yu, Zhibin; Jin, Weidong; and Jiang, Haiying (2020) "Action Recognition by Improved Dense Trajectories," Journal of System Simulation: Vol. 29: Iss. 9, Article 23.
DOI: 10.16182/j.issn1004731x.joss.201709023
Available at: https://dc-china-simulation.researchcommons.org/journal/vol29/iss9/23

First Page

2053

DOI Link

https://doi.org/10.16182/j.issn1004731x.joss.201709023

Last Page

2058

CLC

TP391

Recommended Citation

Xu Peizhen, Yu Zhibin, Jin Weidong, Jiang Haiying. Action Recognition by Improved Dense Trajectories[J]. Journal of System Simulation, 2017, 29(9): 2053-2058.

DOI

10.16182/j.issn1004731x.joss.201709023