Hybrid Human Activity Recognition: Integrating Traditional Feature Engineering with Deep Learning Approach
Hybrid Human Activity Recognition: Integrating Traditional Feature Engineering with Deep Learning Approach
Authors
Vijaya J.
Assistant Professor/ Dept DSAI, IIITNR, Raipur, Chhattisgarh (India)
Nenavathu Pranay
UG Student/ IIITNR, Raipur, Chhattisgarh (India)
G.S. Abhinav
UG Student/ IIITNR, Raipur, Chhattisgarh (India)
Alla Abhiram
UG Student/ IIITNR, Raipur, Chhattisgarh (India)
Bypuneni Chaitanya Krishna
UG Student/ IIITNR, Raipur, Chhattisgarh (India)
Publication Information
Journal Title: International Journal of Research and Scientific Innovation (IJRSI)
Author(s):J.,Vijaya ;Pranay,Nenavathu ;Abhinav,G.S.;Abhiram, Alla ;Krishna,Bypuneni Chaitanya
Published On: 03/10/2026
Volume: 12
Issue: 11
First Page: 1017
Last Page: 1032
ISSN: 2321-2705
- Cite this Article J.,Vijaya ;Pranay,Nenavathu ;Abhinav,G.S.;Abhiram, Alla ;Krishna,Bypuneni Chaitanya :Hybrid Human Activity Recognition: Integrating Traditional Feature Engineering with Deep Learning Approach, Volume 12 Issue 11, International Journal of Research and Scientific Innovation (IJRSI), 1017-1032, Published on 03/10/2026, Available at https://rsisinternational.org/journals/ijrsi/article.php?slug=hybrid-human-activity-recognition-integrating-traditional-feature-engineering-with-deep-learning-approach
Abstract
Human Activity Recognition (HAR) is a vital research area with applications in healthcare, security, and intelligent environments. This paper presents a hybrid framework that combines traditional feature engineering with deep learning to enhance HAR performance. It leverages the Histogram of Oriented Gradients (HoG) for spatial feature extraction and Support Vector Machines (SVM) for structured classification. Additionally, Vision Transformers (ViT) and ResNet architectures are integrated to improve accuracy: ViT captures global dependencies through attention mechanisms, while ResNet enhances deep feature learning through skip connections. Experimental results demonstrate that this approach balances computational efficiency, interpretability, and high accuracy on large datasets.
Keywords:
Human Activity Recognition (HAR)
