Jul 2026· 2026 IEEE International Workshop on Metrology for Living Environment (MetroLivEnv)· pp. 145-150· 0 citations· 21 references
Abstract
Human Activity Recognition (HAR) focuses on measuring human activities from inertial sensor signals acquired through wearable devices. While machine learning and deep learning approaches achieve high classification performance, they often suffer from limited interpretability, high computational cost, and reduced generalization across users. From a measurement perspective, there is therefore increasing interest in identifying simple, physically meaningful descriptors that can be reliably derived from sensor data. In this work, we investigate jerk magnitude, defined as the time derivative of acceleration, as a standalone kinematic measurand for activity discrimination. Triaxial acceleration data were acquired using an Empatica EmbracePlus smartwatch from 20 subjects performing five Activities of Daily Living (ADLs) with different intensity levels. The jerk magnitude was computed through a signal processing pipeline and summarized over short temporal windows using basic statistical features, with particular focus on the mean jerk. To evaluate the discriminative capability of this descriptor, a fully non-parametric statistical framework was adopted, combining the Kruskal–Wallis test for global analysis and Dunn's post-hoc test with Bonferroni correction for pairwise comparisons. The results show that mean jerk magnitude exhibits statistically significant differences across all activity classes $(p<0.001$), enabling clear discrimination between static, low-intensity, and highly dynamic movements. Pairwise analysis confirms strong separability for most activity combinations, while highlighting limitations in distinguishing tasks with similar motion smoothness but different spatial orientation. These findings demonstrate that a single, computationally lightweight and physically interpretable measurand can provide robust activity discrimination without relying on complex models. The proposed approach establishes a reproducible baseline for HAR and highlights the potential of measurement-driven, physics-based descriptors for low-power wearable applications.
Human Activity Recognition (HAR) plays a significant role in various applications, from learning a discipline to physical rehabilitation. In older adults, activity patterns can indicate levels of frailty, which helps inform the design of physical training programs to prevent falls and maintain mobility. HAR sensing ranges from wearable sensors such as IMUs and RGB cameras to video, specialized gait laboratories, perturbation units, VR, and other modalities. This paper presents a comprehensive study of depth-based, skeleton-driven HAR and age group recognition (AGR) using data collected from real-world nursing home environments. Depth sensors offer a privacy-preserving and non-invasive alternative to wearable and RGB-based systems, enabling continuous 24-h monitoring without requiring user compliance. We systematically evaluate multiple modeling paradigms, including classical machine learning models (DT, RF, KNN, SVM, HMM, HMM+SVM), sequence-based models (LSTM, TCN, ARNN), and graph-based approaches, using skeletal joint data extracted from depth images. Experiments are conducted on two heterogeneous datasets: NTU RGB+D (younger adults) and ETAP-DID (older adults). We analyze the impact of different joint subset configurations (full-body, limb-only, leg-only, and torso-only) and compare raw joint representations with handcrafted time-series features (TSFEL) for frame-based HAR. Beyond activity recognition, we introduce an AGR pipeline to distinguish younger from older adults based on skeletal motion patterns. We investigate multiple feature representations, including absolute joint positions, root-relative coordinates, bone vectors, and joint velocities, and provide interpretability through feature importance and saliency analysis to identify age-discriminative joints and motion cues. Our study provides a comprehensive analysis of various HAR models applied to depth data, examining model performance and the contribution of joint-based features to HAR and AGR. Our study highlights the potential for personalized privacy-preserved monitoring and intervention in nursing homes.
R. Paul, Alp Göktug Tanman, Yale Hartmann et al.· Italian National Conference...· 0 citations
The main issues that have been identified by the study include sensor noise, user variability and computational constraints and future prospects of the study is given on context-aware systems and edge intelligence.
Zainab Abdullahi· International Journal of App...· 0 citations
A novel activity intensity score (AIS) framework that provides a nonintrusive and continuous measure of activity intensity by analyzing video data and enables robust, real-time measurement of movement intensity for applications ranging from healthcare and workplace ergonomics to sports analytics and adaptive HVAC control.
Moein Younesi Heravi, Inbae Jeong, Youjin Jang· Journal of computing in civi...· 0 citations
This study reviews pre-2018 developments and proposes an improved HAR framework incorporating advanced feature engineering, sensor fusion, and ensemble learning techniques, which shows strong potential in healthcare, fitness, and smart environments.
Silvia Diallo, F. Z. Idrissi· International Journal of Mod...· 0 citations
This study investigates the impact of different preprocessing techniques on activity recognition accuracy, such as filtering the acceleration to obtain the feature ACCfil, and results indicate that incorporating features such as ACCfil, HRR, and ratio of unfiltered to filtered acceleration improves activity recognition accuracy in some cases.
M. Bermúdez-Edo, Daniel Bolaños-Martinez, Alberto Durán López et al.· International Journal of Int...· 0 citations
Machine learning (ML) algorithms can be used to extract clinically meaningful information from movement data captured by inertial measurement units (IMUs), but many human activity recognition (HAR) pipelines are developed on large laboratory datasets that may not reflect small, heterogeneous, real-world samples. The purpose of this study is to systematically compare the accuracy of multiple ML models, feature sets (both simple and expanded), class balancing strategies, null and transition period handling techniques, and sensor configurations for recognizing a set of four everyday activities extracted from IMU time series data from an age-diverse population. Six ML classifiers were trained and tested: multilayer perceptron, random forest, k-nearest neighbors, logistic regressor, CatBoost, and gaussian naive bayes. These approaches were used in a pipeline with differing sampling techniques including the synthetic minority oversampling technique or random undersampling, and feature handling steps including principal component analysis or a Select-From-Model metatransformer. Additionally, two deep learning methods, DeepConvLSTM and ResGCNN, were trained and tested. Accuracy, precision, recall, and area under the receiver operating characteristic curve were compared to a dummy classifier as a benchmark approximation to chance performance. All pipelines performed better than the dummy classifier, with model accuracy ranging between 0.427 and 0.644. This study demonstrated the ability of several ML algorithms to properly recognize a set of functional activities using limited IMU data from both children and adults.
Hans E. Anderson, Robert A. Scheidt, Kimberly D. Bassindale· Italian National Conference...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
