Open Access iconOpen Access

ARTICLE

crossmark

HRNetO: Human Action Recognition Using Unified Deep Features Optimization Framework

Tehseen Ahsan1,*, Sohail Khalid1, Shaheryar Najam1, Muhammad Attique Khan2, Ye Jin Kim3, Byoungchol Chang4

1 Department of Electrical Engineering, Riphah International University, Peshawar Rd, near Hajj Complex, I-14, Islamabad, Islamabad Capital Territory, 46000, Pakistan
2 Departmnt of Computer Science, HITEC University, Taxila, 47080, Pakistan
3 Department of Computer Science, Hanyang University, Seoul, 04763, Korea
4 Center for Computational Social Science, Hanyang University, Seoul, 04763, Korea

* Corresponding Author: Tehseen Ahsan. Email: email

Computers, Materials & Continua 2023, 75(1), 1089-1105. https://doi.org/10.32604/cmc.2023.034563

Abstract

Human action recognition (HAR) attempts to understand a subject’s behavior and assign a label to each action performed. It is more appealing because it has a wide range of applications in computer vision, such as video surveillance and smart cities. Many attempts have been made in the literature to develop an effective and robust framework for HAR. Still, the process remains difficult and may result in reduced accuracy due to several challenges, such as similarity among actions, extraction of essential features, and reduction of irrelevant features. In this work, we proposed an end-to-end framework using deep learning and an improved tree seed optimization algorithm for accurate HAR. The proposed design consists of a few significant steps. In the first step, frame preprocessing is performed. In the second step, two pre-trained deep learning models are fine-tuned and trained through deep transfer learning using preprocessed video frames. In the next step, deep learning features of both fine-tuned models are fused using a new Parallel Standard Deviation Padding Max Value approach. The fused features are further optimized using an improved tree seed algorithm, and select the best features are finally classified by using the machine learning classifiers. The experiment was carried out on five publicly available datasets, including UT-Interaction, Weizmann, KTH, Hollywood, and IXAMS, and achieved higher accuracy than previous techniques.

Keywords


Cite This Article

APA Style
Ahsan, T., Khalid, S., Najam, S., Khan, M.A., Kim, Y.J. et al. (2023). Hrneto: human action recognition using unified deep features optimization framework. Computers, Materials & Continua, 75(1), 1089-1105. https://doi.org/10.32604/cmc.2023.034563
Vancouver Style
Ahsan T, Khalid S, Najam S, Khan MA, Kim YJ, Chang B. Hrneto: human action recognition using unified deep features optimization framework. Comput Mater Contin. 2023;75(1):1089-1105 https://doi.org/10.32604/cmc.2023.034563
IEEE Style
T. Ahsan, S. Khalid, S. Najam, M.A. Khan, Y.J. Kim, and B. Chang "HRNetO: Human Action Recognition Using Unified Deep Features Optimization Framework," Comput. Mater. Contin., vol. 75, no. 1, pp. 1089-1105. 2023. https://doi.org/10.32604/cmc.2023.034563



cc This work is licensed under a Creative Commons Attribution 4.0 International License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
  • 766

    View

  • 410

    Download

  • 0

    Like

Share Link