Deep Learning in Human Activity Recognition with Wearable Sensors: A Review on Advances

  title={Deep Learning in Human Activity Recognition with Wearable Sensors: A Review on Advances},
  author={Shibo Zhang and Yaxuan Li and Shen Zhang and Farzad Shahabi and Stephen Xia and Y. Deng and Nabil Alshurafa},
  journal={Sensors (Basel, Switzerland)},
Mobile and wearable devices have enabled numerous applications, including activity tracking, wellness monitoring, and human–computer interaction, that measure and improve our daily lives. Many of these applications are made possible by leveraging the rich collection of low-power sensors found in many mobile and wearable devices to perform human activity recognition (HAR). Recently, deep learning has greatly pushed the boundaries of HAR on mobile and wearable devices. This paper systematically… 
Wearable Sensor-Based Human Activity Recognition with Transformer Model
The transformer model, a deep learning neural network model developed primarily for the natural language processing and vision tasks, was adapted for a time-series analysis of motion signals, and the expected future relevance of the transformer model for human activity recognition is suggested.
Explaining One-Dimensional Convolutional Models in Human Activity Recognition and Biometric Identification Tasks
This paper shows how to generate visual explanations about the trained models’ decision making on both HAR and biometric user identification (BUI) tasks and the correlation between them, and adapted gradient-weighted class activation mapping to one-dimensional convolutional neural networks (CNN) architectures to produce visual explanations.
A Pilot Study of the Efficiency of LSTM-Based Motion Classification Algorithms Using a Single Accelerometer
Through suitable combinations of the two factors that constitute deep-learning models according to the data, the potential development and use of efficient models have been verified and the importance of the conditions for hidden units and epochs that are suitable for the given data and the adverse effects of overtraining is determined.
Quantum Water Strider Algorithm with Hybrid-Deep-Learning-Based Activity Recognition for Human–Computer Interaction
This paper presents a new quantum water strider algorithm with hybrid-deep-learning-based activity recognition (QWSA-HDLAR) model for HCI that mainly aims to recognize the different types of activities.
Investigating the Impact of Information Sharing in Human Activity Recognition
This study utilized a public domain dataset that was specifically collected to include variations in smartphone positioning to investigate the effect of information sharing between the training and testing datasets and revealed that XGBoost takes the least computation time while providing high prediction accuracy.
A Low-Cost In-situ System for Continuous Multi-Person Fever Screening
With the recent societal impact of COVID-19, companies and government agencies alike have turned to thermal camera based skin temperature sensing technology to help screen for fever. However, the
Electrical properties modulation of PVA-glycerol based composites for flexible sensors
Physical and electrical properties of films composed of polyvinyl alcohol (PVA) polymer with glycerol (Gly) used as a plasticizer are reported. These plasticized polymer blends are promising for
Wearable Sensor-Based Human Activity Recognition with Hybrid Deep Learning Model
A hybrid deep learning model that amalgamates a one-dimensional Convolutional Neural Network with a bidirectional long short-term memory (1D-CNN-BiLSTM) model is proposed for wearable sensor-based human activity recognition and outshines the existing methods.


Human Activity Recognition With Smartphone and Wearable Sensors Using Deep Learning Techniques: A Review
This review paper focuses on providing profound concise descriptions of deep learning techniques used in smartphone and wearable sensor-based recognition systems, and categorized into conventional and hybrid deep learning models described with its uniqueness, merits, and limitations.
Human Activity Recognition Using Wearable Sensors by Deep Convolutional Neural Networks
This work assembles signal sequences of accelerometers and gyroscopes into a novel activity image, which enables Deep Convolutional Neural Networks (DCNN) to automatically learn the optimal features from the activity image for the activity recognition task.
Deep learning for human activity recognition: A resource efficient implementation on low-power devices
A human activity recognition technique based on a deep learning methodology is designed to enable accurate and real-time classification for low-power wearable devices to obtain invariance against changes in sensor orientation, sensor placement, and in sensor acquisition rates.
Deep Learning Models for Real-time Human Activity Recognition with Smartphones
A smartphone inertial accelerometer-based architecture for HAR is designed and a real-time human activity classification method based on a convolutional neural network (CNN) is proposed, which uses a CNN for local feature extraction on the UCI and Pamap2 datasets.
Deep Learning for Sensor-based Human Activity Recognition
This study presents a survey of the state-of-the-art deep learning methods for sensor-based human activity recognition and proposes a new taxonomy to structure the deep methods by challenges.
Placement Effect of Motion Sensors for Human Activity Recognition using LSTM Network
  • S. Mekruksavanich, A. Jitpattanakul, P. Thongkum
  • Computer Science
    2021 Joint International Conference on Digital Arts, Media and Technology with ECTI Northern Section Conference on Electrical, Electronics, Computer and Telecommunication Engineering
  • 2021
The experimental results show that by using heterogeneous sensor data fusion, the chest position is ideal for physical activity identification with a maximum accuracy of 94.18%.
Real-Time Human Activity Recognition Using Conditionally Parametrized Convolutions on Mobile and Wearable Devices
This paper proposes a computation efficient CNN using conditionally parametrized convolution for real-time HAR on mobile and wearable devices and evaluates the proposed method on four public benchmark HAR datasets, achieving state-of-the-art accuracy without compromising computation cost.
Towards multimodal deep learning for activity recognition on mobile devices
This work investigates the opportunity to use deep learning to perform integration of sensor data from multiple sensors, and initial results with a variant of a Restricted Boltzmann Machine (RBM), show better performance with this new approach compared to classic solutions.
A multibranch CNN-BiLSTM model for human activity recognition using wearable sensor data
A hybrid of convolutional neural network (CNN) and bidirectional long short-term memory (BiLSTM) is used, which does automatic feature extraction from the raw sensor data with minimal data pre-processing and outperforms the other compared approaches.