HYBRID CNN–LSTM-BASED MULTI-BIOMETRIC HUMAN IDENTIFICATION USING FACE AND GAIT
DOI:
https://doi.org/10.29121/shodhkosh.v7.i11s.2026.8268Keywords:
Multi-Biometric Framework, Convolutional Neural Networks (Cnns), Lstm, Biometric Authentication, Hybrid Deep Learning Multi-Biometric Framework (Hdl-Mbf)Abstract [English]
Biometric research has been increasing in response to growing security concerns. Face and gait biometrics are safe, non-invasive, and can be collected anonymously, without the person’s knowledge or consent. These two biometrics are used for a surveillance System. This paper introduces a Hybrid Deep Learning Multi-Biometric Framework (HDL-MBF) for human identification that incorporates both facial and gait features. Convolutional Neural Networks (CNNs) are used to detect discriminative spatial features in face and gait images. LSTM networks, like their counterparts in the Long Short-Term Memory dataset, capture temporal dynamics from face and gait sequences. The PCA method has been applied to extract features, and, in reconstruction, a CNN with LSTM has been employed to increase accuracy over inverse PCA. This pair then combines the strengths of these two methods to achieve reliable, accurate identification. Using the experimental results, which show that the proposed framework achieves 99.89% accuracy in deep-score fusion, it surpasses traditional approaches and single-biometric techniques by significantly reducing processing time. This system is especially suitable for law enforcement, border control, and defence applications and can be easily accessed remotely via drone, making it potentially viable for the highest-security domains. Future research will focus on improving computational efficiency and expanding the framework by incorporating additional biometric modalities to enhance adaptability and robustness, as well as reducing time complexity
References
A. Kale, A. K. Roychowdhury, and R. Chellappa, “Fusion of gait and face for human identification,” in 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing, IEEE, pp. V-901–4. doi: 10.1109/ICASSP.2004.1327257.
A. Kumar, S. Jain, and M. Kumar, “Comparative Study of Multi-Biometrics Authentication Using Machine Learning Algorithms,” in 2024 11th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO), IEEE, Mar. 2024, pp. 1–5. doi: 10.1109/ICRITO61523.2024.10522125.
A. Kumar, S. Jain, and M. Kumar, “Deep Learning based Fusion for a Multi-Biometric Identification Using LSTM,” in 2024 1st International Conference on Advanced Computing and Emerging Technologies (ACET), IEEE, Aug. 2024, pp. 1–6. doi: 10.1109/ACET61898.2024.10730213.
A. Kumar, S. Jain, and M. Kumar, “Face and gait biometrics authentication system based on simplified deep neural networks,” International Journal of Information Technology, vol. 15, no. 2, pp. 1005–1014, Feb. 2023, doi: 10.1007/s41870-022-01087-5.
A. Mahajan and S. K. Singla, “DeepBio: A Deep CNN and Bi-LSTM Learning for Person Identification Using Ear Biometrics,” Computer Modeling in Engineering & Sciences, vol. 141, no. 2, pp. 1623–1649, 2024, doi: 10.32604/cmes.2024.054468.
A. Prakash, S. Thejaswin, A. Nambiar, and A. Bernardino, “Adapt-FuseNet: Context-aware Multimodal Adaptive Fusion of Face and Gait Features using Attention Techniques for Human Identification,” in 2023 IEEE International Joint Conference on Biometrics (IJCB), IEEE, Sep. 2023, pp. 1–10. doi: 10.1109/IJCB57857.2023.10448765.
B. A. El-Rahiem, M. Amin, A. Sedik, F. E. A. El Samie, and A. M. Iliyasu, “An efficient multi-biometric cancellable biometric scheme based on deep fusion and deep dream,” J Ambient Intell Humaniz Comput, vol. 13, no. 4, pp. 2177–2189, Apr. 2022, doi: 10.1007/s12652-021-03513-1.
C. Chen, Z. Du, L. He, Y. Shi, J. Wang, and W. Dong, “A Novel Gait Pattern Identification Method Based on LSTM-CNN for Lower Limb Exoskeleton,” J Bionic Eng, vol. 18, no. 5, pp. 1059–1072, Sep. 2021, doi: 10.1007/s42235-021-00083-y.
C. Shen, S. Yu, J. Wang, G. Q. Huang, and L. Wang, “A Comprehensive Survey on Deep Gait Identification: Algorithms, Datasets, and Challenges,” IEEE Trans Biom Behav Identity Sci, pp. 1–1, 2024, doi: 10.1109/TBIOM.2024.3486345.
D. M. Ahmed and B. Sh. Mahmood, “Integration of Face and Gait Identification via Transfer Learning: A Multiscale Biometric Identification Approach,” Traitement du Signal, vol. 40, no. 5, pp. 2179–2190, Oct. 2023, doi: 10.18280/ts.400535.
F. et al. Liu, “Person Identification at Altitude and Range: Fusion of Face, Body Shape and Gait,” 2025.
Face dataset CASIA-WebFace | Kaggle
G. Shakhnarovich, L. Lee, and T. Darrell, “Integrated face and gait identification from multiple views,” in Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Identification. CVPR 2001, IEEE Comput. Soc, pp. I-439-I–446. doi: 10.1109/CVPR.2001.990508.
Gait data set The dataset is available on the open-source CASIA-B dataset.
http://www.cbsr.ia.ac.cn/english/Gait%20Databases.aspx
I. U. Khan, S. Afzal, and J. W. Lee, “Human Activity Identification via Hybrid Deep Learning Based Model,” Sensors, vol. 22, no. 1, p. 323, Jan. 2022, doi: 10.3390/s22010323.
J Zhai, Y Xu, and X Yan, “Heterologous Image Matching Based on Salience Region,” IAENG Int J Comput Sci, 2025.
M. et al. Wang, “DyGait: Exploiting Dynamic Representations for High-performance Gait Identification,” Proceedings of the IEEE/CVF international conference on computer vision. 2023., 2023.
M. Ramzan and S. Dawn, “Fused CNN-LSTM deep learning emotion identification model using electroencephalography signals,” International Journal of Neuroscience, vol. 133, no. 6, pp. 587–597, Jun. 2023, doi: 10.1080/00207454.2021.1941947.
Moon, H., Bey, O., Boubezoul, A., Oukhellou, L., & Mohammed, S. (2025). Real-Time LSTM-Driven Dynamic Gait Mode Detection for Enhanced Control of Actuated-Ankle-Foot Orthosis. IEEE Transactions on Robotics.
N. Senthilkumar, M. Manimegalai, S. Karpakam, S. R. Ashokkumar, and M. Premkumar, “Human action identification based on spatial–temporal relational model and LSTM-CNN framework,” Mater Today Proc, vol. 57, pp. 2087–2091, 2022, doi: 10.1016/j.matpr.2021.12.004.
P. M, R. Naidu M, and A. P, “Combining Deep Learning Techniques for Enhanced Human Activity Identification: A Hybrid CNN-LSTM Fusion Approach,” in 2024 IEEE International Conference on Contemporary Computing and Communications (InC4), IEEE, Mar. 2024, pp. 1–7. doi: 10.1109/InC460750.2024.10649335.
R. Liao, Z. Li, S. S. Bhattacharyya, and G. York, “PoseMapGait: A model-based gait identification method with pose estimation maps and graph convolutional networks,” Neurocomputing, vol. 501, pp. 514–528, Aug. 2022, doi: 10.1016/j.neucom.2022.06.048.
Ravina Gupta, Sarika Jain, and Manoj Kumar, “Thermal Vision for Airfield Safety: A Dynamic FOD Detection Framework,” IAENG Int J Comput Sci, vol. 52, no. 4, pp. 1178–1186, 2025.
S. A. F. Manssor, S. Sun, and M. A. M. Elhassan, “Real-Time Human Identification at Night via Integrated Face and Gait Identification Technologies,” Sensors, vol. 21, no. 13, p. 4323, Jun. 2021, doi: 10.3390/s21134323.
S. Naoui, M. H. Elhdhili and L. A. Saidane, "Novel Smart Home Authentication Protocol LRP-SHAP," 2019 IEEE Wireless Communications and Networking Conference (WCNC), Marrakesh, Morocco, 2019, pp. 1-6, doi: 10.1109/WCNC.2019.8885493
S. Salturk and N. Kahraman, “Deep learning-powered multimodal biometric authentication: integrating dynamic signatures and facial data for enhanced online security,” Neural Comput Appl, vol. 36, no. 19, pp. 11311–11322, Jul. 2024, doi: 10.1007/s00521-024-09690-2.
T. S, A. Prakash, A. Nambiar, and A. Bernadino, “Exploring Fusion Techniques and Explainable AI on Adapt-FuseNet: Context-Adaptive Fusion of Face and Gait for Person Identification,” IEEE Trans Biom Behav Identity Sci, vol. 6, no. 4, pp. 515–527, Oct. 2024, doi: 10.1109/TBIOM.2024.3405081.
V. Rajasekar et al., “Enhanced multimodal biometric identification approach for smart cities based on an optimized fuzzy genetic algorithm,” Sci Rep, vol. 12, no. 1, p. 622, Jan. 2022, doi: 10.1038/s41598-021-04652-3.
V. Rani and M. Kumar, “Human gait identification: A systematic review,” Multimed Tools Appl, vol. 82, no. 24, pp. 37003–37037, Oct. 2023, doi: 10.1007/s11042-023-15079-5.
V. Sharma, M. Gupta, A. Kumar, and D. Mishra, “Video Processing Using Deep Learning Techniques: A Systematic Literature Review,” IEEE Access, vol. 9, pp. 139489–139507, 2021, doi: 10.1109/ACCESS.2021.3118541.
W. Meng et al., “Emotion identification via affective EEG signals: State of the art,” Neurocomputing, vol. 643, p. 130418, Aug. 2025, doi: 10.1016/j.neucom.2025.130418.
X. Wang and W. Q. Yan, “Human Gait Identification Based on Frame-by-Frame Gait Energy Images and Convolutional Long Short-Term Memory,” Int J Neural Syst, vol. 30, no. 01, p. 1950027, Jan. 2020, doi: 10.1142/S0129065719500278.
Y. Ojha, “Artificial Intelligence in Armed Conflict: Perspectives from International Humanitarian Law,” SSRN Electronic Journal, 2025, doi: 10.2139/ssrn.5162209.
Z. W. Wenrui Shen, “Person Re-Identification Algorithm Based on Improved ResNet,” IAENG International Journal of Applied Mathematics, vol. 54, no. 5, 2024.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Amit Kumar, Sarika Jain, Manoj Kumar
This work is licensed under a Creative Commons Attribution 4.0 International License.
With the licence CC-BY, authors retain the copyright, allowing anyone to download, reuse, re-print, modify, distribute, and/or copy their contribution. The work must be properly attributed to its author.
It is not necessary to ask for further permission from the author or journal board.
This journal provides immediate open access to its content on the principle that making research freely available to the public supports a greater global exchange of knowledge.
