An Efficient Human Activity Recognition Technique Based on Deep Learning

• Published in: Pattern recognition and image analysis Vol. 29; no. 4; pp. 702 - 715
• Main Authors: Khelalef, A., Ababsa, F., Benoudjit, N.
• Format: Journal Article
• Language: English
• Published: Moscow Pleiades Publishing 01.10.2019; Springer Nature B.V; MAIK Nauka/Interperiodica (МАИК Наука/Интерпериодика)
• Subjects: Applied Problems; Artificial Intelligence; Artificial neural networks; Computer Science; Computer Vision and Pattern Recognition; Deep learning; Feature extraction; Human activity recognition; Image Processing and Computer Vision; Machine learning; Pattern Recognition; Video data; deep learning; features extraction; convolutional neural network (CNN); classification; human activity recognition; Deep learning; Convolutional neural network (CNN); Human activity recognition; Features extraction; Classification
• ISSN: 1054-6618; 1555-6212
• DOI: 10.1134/S1054661819040084

In this paper, we present a new deep learning-based human activity recognition technique. First, we track and extract human body from each frame of the video stream. Next, we abstract human silhouettes and use them to create binary space-time maps (BSTMs) which summarize human activity within a defined time interval. Finally, we use convolutional neural network (CNN) to extract features from BSTMs and classify the activities. To evaluate our approach, we carried out several tests using three public datasets: Weizmann, Keck Gesture and KTH Database. Experimental results show that our technique outperforms conventional state-of-the-art methods in term of recognition accuracy and provides comparable performance against recent deep learning techniques. It’s simple to implement, requires less computing power, and can be used for multi-subject activity recognition.