Title: Advancing human action recognition: wavelet-DTW enhanced deep learning with multi-head attention

Authors: Salma Tayeb; Hafssa Mediani; Soufiana Mekouar; Mohammed Majid Himmi

Addresses: Laboratory of Computer Science, Applied Mathematics, Artificial Intelligence and Pattern Recognition (LIMIARF), Department of Physics, Faculty of Sciences, Mohammed V University in Rabat, Rabat, Morocco ' Laboratory of Computer Science, Applied Mathematics, Artificial Intelligence and Pattern Recognition (LIMIARF), Department of Physics, Faculty of Sciences, Mohammed V University in Rabat, Rabat, Morocco ' Department of Geomorphology and Geomatics, Scientific Institute, Mohammed V University in Rabat, Rabat, Morocco ' Laboratory of Computer Science, Applied Mathematics, Artificial Intelligence and Pattern Recognition (LIMIARF), Department of Physics, Faculty of Sciences, Mohammed V University in Rabat, Rabat, Morocco

Abstract: This study introduces a novel approach to human action recognition by combining discrete wavelet transform (DWT) for multi-scale feature extraction, dynamic time warping (DTW) for sequence alignment, and multi-head attention (MHA) within a convolutional bidirectional long-short-term memory (Conv-BiLSTM) framework. This integration enables precise recognition across diverse temporal scales and complex motion dynamics, setting our approach apart from traditional models. The framework effectively addresses challenges such as imbalanced class distributions and varying action speeds in short video clips, optimising both accuracy and computational efficiency. On benchmark datasets UCF101 and HMDB51, the model achieves 97.02% and 91.6% accuracy, respectively, outperforming current state-of-the-art methods. Statistical analyses and ablation studies demonstrate the contribution of each component to the model's performance. A detailed comparison with other methods highlights its advantages for real-time applications. This work advances human action recognition by combining traditional and modern techniques in an optimised, low-cost architecture suitable for dynamic environments.

Keywords: human action recognition; HAR; discrete wavelet transform; DWT; dynamic time warping; DTW; convolutional and bidirectional long short-term memory; Conv-BiLSTM; multi-head attention; MHA.

DOI: 10.1504/IJICA.2025.145027

International Journal of Innovative Computing and Applications, 2025 Vol.15 No.2, pp.102 - 117

Received: 23 Jul 2024
Accepted: 04 Jan 2025
Published online: 17 Mar 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article