Nov 2026· Journal of computing in civil engineering· Vol 40· 0 citations· 39 references
Computer Science
TL;DR
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.
Abstract
Accurate real-time estimation of human activity intensity is essential for diverse applications such as health monitoring, ergonomics, sports science, and adaptive building management. However, existing methods often depend on intrusive wearable sensors, discrete activity classifications, or extensive training datasets, which compromise their practicality and generalizability. To address these gaps, we propose a novel activity intensity score (AIS) framework that provides a nonintrusive and continuous measure of activity intensity by analyzing video data. The proposed method applies pose estimation to video data to extract body landmarks, which are then used to compute kinematic parameters including the angular velocity, angular acceleration, range of motion, peak speed, movement frequency, and rotational energy across defined kinematic chains (e.g., arms, legs, torso). These parameters are then normalized and combined through an optimized weighted summation to produce a continuous activity intensity metric. Experimental validation was conducted with 20 participants performing various activities from low, moderate, and high intensity. Results demonstrated strong correlations between AIS scores and both activity intensity levels (Spearman’s
ρ
=
0.943
,
p
<
0.001
) and participants perceived exertion ratings (Pearson’s
r
=
0.923
,
p
<
0.001
). Statistical comparisons demonstrated that the AIS values effectively discriminate among these three intensity categories (Spearman’s
ρ
=
0.943
) and significant group differences confirmed by ANOVA (
p
<
0.001
). Moreover, the AIS exhibited a strong correlation (Pearson’s
r
=
0.923
) with self-reported exertion (Borg rating of perceived exertion), indicating consistency with participants’ subjective perceptions. This occupant-invariant and domain-independent method enables robust, real-time measurement of movement intensity for applications ranging from healthcare and workplace ergonomics to sports analytics and adaptive HVAC control.
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
Human motion perception is an important function in intelligent electromechanical systems, especially in human-computer interaction, rehabilitation training, and safety monitoring. Existing skeleton-based deep learning methods can model spatial and temporal motion patterns, but they usually require large datasets and relatively high computational resources. This limits their use in small-sample and lightweight deployment scenarios. This study examines whether compact temporal features extracted from skeleton coordinates can be used for efficient correctness detection of human movements. A cross-subject skeleton motion dataset containing 2,589 samples was used, and the task was to classify actions as correct or incorrect. Three coordinate-based feature representations were compared: flattened original XY coordinates, flattened standardized XY coordinates, and XY- based temporal statistical features. A graph-based ST-GCN model was also introduced as a deep learning baseline. The results show that the XY temporal statistical features achieved the best overall performance with only 400 features, reaching a balanced accuracy of 0.8061 and a macro F1 score of 0.7985. Under the same evaluation conditions, this method outperforms the two coordinate flattening representation methods as well as the ST-GCN baseline model. These findings indicate that lightweight statistical features can provide an efficient and interpretable method for motion accuracy assessment in electromechanical systems.
For structural dynamics and health monitoring, vision-based vibration measurement has become an attractive approach thanks to its full-field, non-contact, and cost-effective aspect. However, conventional frame-based cameras are inherently constrained by frame rate, motion blur, and exposure duration, while producing vast amounts of redundant data. These limitations restrict their ability to capture high-frequency vibrations over extended periods or under low illumination.
In contrast, event cameras asynchronously record pixel-level brightness changes with microsecond latency and a high dynamic range. This enables high-frequency motion tracking under ambient light or uneven light condition and ensures efficient data use, as events are generated only where motion occurs.
This study presents a unified event-based framework for estimating vibration and modal parameters within a dynamic state-space formulation. Displacement information is extracted from images reconstructed from event streams, while velocity information is obtained directly from the events. By combining these complementary motion signals with a physics-based vibration model via data assimilation, the method enables accurate and continuous estimation of structural vibrations with strong robustness to noise.
Experiments were performed on a cantilever beam with an event camera and a reference accelerometer synchronized in operation, including both free and forced vibration tests. In the free-vibration tests, modal parameters were identified from the decaying response, whereas the forced-vibration setup used a shaker to apply harmonic excitation, enabling the estimation of the operational deflection shapes under steady-state conditions. The extracted modal properties closely matched the accelerometer measurements, validating the proposed approach.
The proposed framework demonstrates the capability of the event-camera-based approach to measure structural vibrations with high temporal fidelity and strong robustness to noise. It combines event data with a physics-based vibration model through data assimilation. This integration enables accurate modal identification and efficient motion estimation without the motion blur, saturation, or illumination constraints inherent to frame-based systems. These findings highlight the potential of event cameras as a robust and data-efficient sensing modality for continuous, high-frequency monitoring in structural dynamics and health assessment.
Yunhyeok Han, V. Yaghoubi· e-Journal of Nondestructive...· 0 citations
Evaluation of the effectiveness of rehabilitation therapy is often based on patient outcomes and functional improvement over time. However, traditional approaches are subjective and heavily influenced by expert perspectives and perceptions, which can lead to inconsistencies in research. This study developed a motion measurement system that improves body positioning by using computer vision techniques to quantitatively assess elbow flexibility and range of motion during rehabilitation. The system uses a novel posture estimation model that analyses video input to identify key body points and match body position, providing smooth, real-time feedback. In addition, the system is designed to be easy to use and simple, so it can be implemented in both clinical settings and home-based rehabilitation programs. Data validation is carried out using a reference dataset that has previously been analyzed, with an absolute mean error of approximately 5.00 degrees and a correlation coefficient of approximately 0.95. The results show that this system can be used to maintain one's own behavior. This technology supports the rehabilitation process by providing clear data, helping medical professionals and health care providers better assess patient progress and make more appropriate decisions. In addition, this technology can improve tele-rehabilitation services, especially for patients who cannot access medical facilities. With that in mind, limitations such as model issues and video quality variations can affect the accuracy of body-position determination.
Kok Swee Sim, Mahda Laina Arnumukti· International Journal on Adv...· 0 citations
Precise biomechanical assessment of athletic movements is essential for intelligent sports analysis and wearable sensing applications. This study presents a multimodal deep learning framework that integrates computer vision, RGB video, depth images, and inertial measurement data to achieve markerless motion capture and quantitative biomechanical evaluation. Through spatiotemporal feature alignment and attention-based multimodal fusion, the proposed system accurately estimates joint kinematics, ground reaction forces, torque distribution, and muscle-related motion characteristics while enabling real-time action recognition and technical assessment. Experimental results demonstrate that the framework achieves a keypoint detection accuracy of 89.4%, an F1 score of 92.3% for complex action recognition, and a root mean square error of 4.2◦ in joint angle estimation, providing reliable support for sports performance optimization and injury risk analysis. The proposed multimodal sensing strategy offers an effective solution for integrating visual perception with wearable sensor information and provides valuable methodological insights for intelligent signal acquisition, wireless sensing, and human-centered electromagnetic monitoring systems in next-generation smart engineering applications.
X. J. Tian, F. L. Geng· Advanced Electromagnetics· 0 citations
The accurate assessment of movement quality in rehabilitation training is a critical challenge in both clinical and home-based settings. Traditional methods often rely on subjective evaluations or limited sensor data, which can lead to inconsistent and inaccurate assessments. This paper introduces a novel methodology that integrates surface electromyography (sEMG) signals with RGB-D video data to provide a comprehensive, real-time analysis of rehabilitation movement quality. The proposed framework leverages the Manifold-driven Event Forecaster, a sophisticated model that captures the intricate biomechanical and neuromuscular aspects of human motion. The methodology is composed of three primary components: a Nonlinear Constraint Optimizer that aligns multimodal data onto a shared manifold, an Agent-based Temporal Segmenter that partitions data into meaningful temporal segments, and a Probabilistic Quality Predictor that quantifies movement quality with an uncertainty-aware approach. The framework employs advanced strategies such as manifold alignment refinement and uncertainty-aware modeling to enhance robustness and interpretability. These innovations enable the system to make accurate, real-time predictions of movement quality, effectively addressing challenges such as variability in patient performance and sensor noise. Experimental results demonstrate the efficacy of the proposed approach, showing significant improvements in the precision and reliability of movement quality assessments. This comprehensive solution has the potential to transform rehabilitation practices by providing objective, data-driven insights into patient progress and therapy effectiveness, ultimately contributing to improved patient outcomes and more personalized rehabilitation programs.
Xiaolu Li, Rong Li, Ruisheng Wu· International Journal of Ima...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
