Abstract
Human activity recognition (HAR) plays a vital role in the fields like security, health analysis, gaming, video streaming, surveillances, etc. Many applications were developed based on HAR using video due to its user-friendly nature and affordable cost. We propose the fusion of convolutional layer and Long-Short-Term Memory deep architecture for HAR. In this model, three convolution layer, two BiLSTM layer is used. Convolution layer extracts local features of the frame data. Extracted features were flattened and feed into BiLSTM layers to process the frame sequence and handles the time dependencies. The performance of the model is evaluated using two public datasets namely UCF101 dataset and HMDB-51 dataset and obtained the average accuracy of 96% and 95%, respectively. Influence of hyperparameter was analysed and tuned parameter is used for implementation. The proposed model provides the better accuracy when compared with other deep learning architectures used for HAR using video.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Y. Wu, Y. Su, R. Feng, N. Yu, X. Zang, Wearable-sensor-based pre-impact fall detection system with a hierarchical classifier. Measurement 140, 283–292 (2019)
J. Zhang, Y. Cao, M. Qiao, L. Ai, K. Sun, Q. Mi, Q. Wang et al., Human motion monitoring in sports using wearable graphene-coated fiber sensors. Sens. Actuators A 274, 132–140 (2018)
M.O. Gani, T. Fayezeen, R.J. Povinelli, R.O. Smith, M. Arif, A.J. Kattan, S.I. Ahamed, A light weight smartphone based human activity recognition system with high accuracy. J. Netw. Comput. Appl. (2019)
W. Jiang, Z. Yin, Human activity recognition using wearable sensors by deep convolutional neural networks, in Proceedings of the 23rd ACM International Conference on Multimedia - MM ’15 (2015)
A. Jain, V. Kanhangad, Human activity classification in smartphones using accelerometer and gyroscope sensors. IEEE Sens. J. 18(3), 1169–1177 (2018)
Y.-L. Hsu, S.-C. Yang, H.-C. Chang, H.-C. Lai, Human daily and sport activity recognition using a wearable inertial sensor network. IEEE Access 6, 31715–31728 (2018)
S.-I. Chu, B.-H. Liu, N.-T. Nguyen, Secure AF relaying with efficient partial relay selection scheme. Int J Commun Syst. 32, e4105 (2019). https://doi.org/10.1002/dac.4105
M. Balaanand, N. Karthikeyan, S. Karthik, R. Varatharajan, G. Manogaran, C.B. Sivaparthipan, An enhanced graph-based semi-supervised learning algorithm to detect fake users on Twitter. J. Supercomput. 75(9), 6085–6105 (2019). https://doi.org/10.1007/s11227-019-02948-w
Y. Tian, X. Wang, L. Chen, Z. Liu, Wearable sensor-based human activity recognition via two-layer diversity-enhanced multiclassifier recognition method. Sensors 19(9), 2039 (2019)
C.A. Ronao, S.-B. Cho, Human activity recognition with smartphone sensors using deep learning neural networks. Expert Syst. Appl. 59, 235–244 (2016)
M.-O. Mario, Human activity recognition based on single sensor square hv acceleration images and convolutional neural networks. IEEE Sen. J. 1–1 (2018)
Y. Shi, Y. Tian, Y. Wang, T. Huang, Sequential deep trajectory descriptor for action recognition with three-stream CNN. IEEE Trans. Multimedia 19(7), 1510–1520 (2017)
J. Zhao, X. Mao, L. Chen, Speech emotion recognition using deep 1D & 2D CNN LSTM networks. Biomed. Signal Process. Control 47, 312–323 (2019)
S. Bai, H. Tang, S. An, Coordinate CNNs and LSTMs to Categorize Scene Images with Multi-views and Multi-levels of Abstraction. Expert Systems with Applications (2018)
M. Wang, Y.D. Zhang, G. Cui, Human motion recognition exploiting radar with stacked recurrent neural network. Digital Signal Processing (2019)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Nithyakani, P., Ferni Ukrit, M. (2022). Video Based Human Gait Activity Recognition Using Fusion of Deep Learning Architectures. In: Manogaran, G., Shanthini, A., Vadivu, G. (eds) Proceedings of International Conference on Deep Learning, Computing and Intelligence. Advances in Intelligent Systems and Computing, vol 1396. Springer, Singapore. https://doi.org/10.1007/978-981-16-5652-1_51
Download citation
DOI: https://doi.org/10.1007/978-981-16-5652-1_51
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-16-5651-4
Online ISBN: 978-981-16-5652-1
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)