Deep, Convolutional, and Recurrent Models for Human Activity Recognition using Wearables

Abstract

Human activity recognition (HAR) in ubiquitous computing is beginning to adopt deep learning to substitute for well-established analysis techniques that rely on hand-crafted feature extraction and classification techniques. From these isolated applications of custom deep architectures it is, however, difficult to gain an overview of their suitability for problems ranging from the recognition of manipulative gestures to the segmentation and identification of physical activities like running or ascending stairs. In this paper we rigorously explore deep, convolutional, and recurrent approaches across three representative datasets that contain movement data captured with wearable sensors. We describe how to train recurrent approaches in this setting, introduce a novel regularisation approach, and illustrate how they outperform the state-of-the-art on a large benchmark dataset. Across thousands of recognition experiments with randomly sampled model configurations we investigate the suitability of each model for different tasks in HAR, explore the impact of hyperparameters using the fANOVA framework, and provide guidelines for the practitioner who wants to apply deep learning in their problem setting.

Extended version has been accepted for publication at International Joint Conference on Artificial Intelligence (IJCAI)

References in corpus (3)

• LSTM: A Search Space Odyssey
• DRAW: A Recurrent Neural Network For Image Generation
• Convolutional Neural Network for Stereotypical Motor Movement Detection in Autism

Cited by in corpus (54)

• Deep Learning for Sensor-based Activity Recognition: A Survey
• Ensembles of Deep LSTM Learners for Activity Recognition using Wearables
• Human Activity Recognition using Inertial, Physiological and Environmental Sensors: a Comprehensive Survey
• Multi-task Self-Supervised Learning for Human Activity Detection
• Deep Learning for Sensor-based Human Activity Recognition: Overview, Challenges and Opportunities
• Sensor-based Gait Parameter Extraction with Deep Convolutional Neural Networks
• Layer-wise training convolutional neural networks with smaller filters for human activity recognition using wearable sensors
• SelfHAR: Improving Human Activity Recognition through Self-training with Unlabeled Data
• Attention-Based Deep Learning Framework for Human Activity Recognition with User Adaptation
• Asymmetric Residual Neural Network for Accurate Human Activity Recognition
• Exploring Contrastive Learning in Human Activity Recognition for Healthcare
• Sequential Weakly Labeled Multi-Activity Localization and Recognition on Wearable Sensors using Recurrent Attention Networks
• DFTerNet: Towards 2-bit Dynamic Fusion Networks for Accurate Human Activity Recognition
• Human Activity Recognition from Wearable Sensor Data Using Self-Attention
• Recognizing Abnormal Heart Sounds Using Deep Learning
• Semi-supervised Federated Learning for Activity Recognition
• Deep Learning-Based Gait Recognition Using Smartphones in the Wild
• Compressing RNNs for IoT devices by 15-38x using Kronecker Products
• edBB: Biometrics and Behavior for Assessing Remote Education
• Modeling Long- and Short-Term Temporal Patterns with Deep Neural Networks
• Run-Time Efficient RNN Compression for Inference on Edge Devices
• Attributes' Importance for Zero-Shot Pose-Classification Based on Wearable Sensors
• A deep learning approach for lower back-pain risk prediction during manual lifting
• Ultra-compact Binary Neural Networks for Human Activity Recognition on RISC-V Processors
• Wearable affect and stress recognition: A review
• Chronic-Pain Protective Behavior Detection with Deep Learning
• Wearable Sensor Data Based Human Activity Recognition using Machine Learning: A new approach
• DanHAR: Dual Attention Network For Multimodal Human Activity Recognition Using Wearable Sensors
• Similarity Embedding Networks for Robust Human Activity Recognition
• Actigraphy-based Sleep/Wake Pattern Detection using Convolutional Neural Networks
• IMUTube: Automatic Extraction of Virtual on-body Accelerometry from Video for Human Activity Recognition
• Deep Learning for Multi-Scale Changepoint Detection in Multivariate Time Series
• Spectrum-Guided Adversarial Disparity Learning
• Topological Persistence Guided Knowledge Distillation for Wearable Sensor Data
• Cross-Country Skiing Gears Classification using Deep Learning
• Session Analysis using Plan Recognition
• Learning Bodily and Temporal Attention in Protective Movement Behavior Detection
• The Benefit of the Doubt: Uncertainty Aware Sensing for Edge Computing Platforms
• Reducing Label Dependency in Human Activity Recognition with Wearables: From Supervised Learning to Novel Weakly Self-Supervised Approaches
• Valid and Exact Statistical Inference for Multi-dimensional Multiple Change-Points by Selective Inference
• Leveraging Activity Recognition to Enable Protective Behavior Detection in Continuous Data
• Stochastic Gradient Descent with Hyperbolic-Tangent Decay on Classification
• Learning behavioral context recognition with multi-stream temporal convolutional networks
• Multi-Modal Recognition of Worker Activity for Human-Centered Intelligent Manufacturing
• MemNet: Memory-Efficiency Guided Neural Architecture Search with Augment-Trim learning
• Hierarchical Self Attention Based Autoencoder for Open-Set Human Activity Recognition
• Multi-Modal Subjective Context Modelling and Recognition
• Lightweight Transformer in Federated Setting for Human Activity Recognition
• Human Activity Recognition using Attribute-Based Neural Networks and Context Information
• rIoT: Enabling Seamless Context-Aware Automation in the Internet of Things
• A Closer Look at Mobile App Usage as a Persistent Biometric: A Small Case Study
• Entropy Decision Fusion for Smartphone Sensor based Human Activity Recognition
• An active smartphone authentication method based on daily cyclical activity
• Towards Reliable, Automated General Movement Assessment for Perinatal Stroke Screening in Infants Using Wearable Accelerometers