Smart Handheld Based Human Activity Recognition Using Multiple Instance Multiple Label Learning
- PDF / 1,905,390 Bytes
- 21 Pages / 439.37 x 666.142 pts Page_size
- 17 Downloads / 232 Views
Smart Handheld Based Human Activity Recognition Using Multiple Instance Multiple Label Learning Jayita Saha1 · Dip Ghosh2 · Chandreyee Chowdhury3 · Sanghamitra Bandyopadhyay2 Accepted: 29 October 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Human activity recognition (HAR) and monitoring is beneficial for many medical applications, such as eldercare and post-trauma rehabilitation after surgery. HAR models based on smartphone’s accelerometer data could provide a convenient and ubiquitous solution to this problem. However, such models are mostly concerned with identifying basic activities such as ‘stand’/‘walk’ and thus the high-level context such as ‘walk in a queue’ for which a set of specific activities is performed remain unnoticed. Consequently, in this paper, we design a HAR framework that can identify a group of activities (rather than a single basic activity) being performed in a time window, thus, enables us to extract more meaningful information about the subject’s overall context. An algorithm is designed to formulate HAR as a multi-instance multi-label (MIML) learning problem. The procedure of generating feature bags of consecutive activity traces having multiple labels is formulated. In this work, the temporal relationship among activities is exploited to obtain a more comprehensive HAR model. Interestingly, the framework is found to completely/partially identify activity sequences that may not even be present in the training dataset. The framework is implemented and found to be working adequately when tested with real dataset collected from 8 users for 12 different activity combinations. MIML-kNN is found to provide maximum average precision (around 90%) even for an unseen test data-set. Keywords Activity recognition · Composite activity · Accelerometer · Semi supervised learning · Multi-instance and multi-label learning * Jayita Saha [email protected] Dip Ghosh [email protected] Chandreyee Chowdhury [email protected] Sanghamitra Bandyopadhyay [email protected] 1
Department of Artificial Intelligence and Data Science, Koneru Lakshmaiah Education Foundation, Deemed to be University, Hyderabad, India
2
Machine Intelligence Unit, Indian Statistical Institute, Kolkata, India
3
Department of CSE, Jadavpur University, Kolkata, India
13
Vol.:(0123456789)
J. Saha et al.
1 Introduction A Human Activity Recognition (HAR) framework both identifies and analyses a person’s daily activities to provide context-aware feedback for healthcare and elderly care [1, 2]. Most of the existing HAR systems recognize primitive classes of daily activities such as sit, stand, walk etc. considering state-of-the-art classifiers that may not evidently give us information about the subject’s context in [3]. The systems have been designed in a way in which they can predict one single activity based on a window of data that was being performed at that particular time instant as considered in [4, 5]. But daily activities are composite and many of them could be
Data Loading...
