eprintid: 27968 rev_number: 8 eprint_status: archive userid: 2 dir: disk0/00/02/79/68 datestamp: 2026-03-30 21:37:32 lastmod: 2026-03-30 21:37:33 status_changed: 2026-03-30 21:37:32 type: article metadata_visibility: show creators_name: Raza, Muhammad Amjad creators_name: Mehmood, Nasir creators_name: Siddiqui, Hafeez Ur Rehman creators_name: Saleem, Adil Ali creators_name: Álvarez, Roberto Marcelo creators_name: Miró Vera, Yini Airet creators_name: Díez, Isabel de la Torre creators_id: creators_id: creators_id: creators_id: creators_id: roberto.alvarez@uneatlantico.es creators_id: yini.miro@uneatlantico.es creators_id: title: Human Activity Recognition in Domestic Settings Based on Optical Techniques and Ensemble Models ispublished: pub subjects: uneat_eng divisions: uneatlantico_produccion_cientifica divisions: unincol_produccion_cientifica divisions: uninimx_produccion_cientifica divisions: uninipr_produccion_cientifica divisions: unic_produccion_cientifica divisions: uniromana_produccion_cientifica full_text_status: public keywords: deep learning; human activity recognition; LSTM; PoseNet; skeleton-based recognition; smart home; Transformer abstract: Human activity recognition (HAR) is essential in many applications, such as smart homes, assisted living, healthcare monitoring, rehabilitation, physiotherapy, and geriatric care. Conventional methods of HAR use wearable sensors, e.g., acceleration sensors and gyroscopes. However, they are limited by issues such as sensitivity to position, user inconvenience, and potential health risks with long-term use. Optical camera systems that are vision-based provide an alternative that is not intrusive; however, they are susceptible to variations in lighting, intrusions, and privacy issues. The paper uses an optical method of recognizing human domestic activities based on pose estimation and deep learning ensemble models. The skeletal keypoint features proposed in the current methodology are extracted from video data using PoseNet to generate a privacy-preserving representation that captures key motion dynamics without being sensitive to changes in appearance. A total of 30 subjects (15 male and 15 female) were sampled across 2734 activity samples, including nine daily domestic activities. There were six deep learning architectures, namely, the Transformer (Transformer), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Multilayer Perceptron (MLP), One-Dimensional Convolutional Neural Network (1D CNN), and a hybrid Convolutional Neural Network–Long Short-Term Memory (CNN–LSTM) architecture. The results on the hold-out test set show that the CNN–LSTM architecture achieves an accuracy of 98.78% within our experimental setting. Leave-One-Subject-Out cross-validation further confirms robust generalization across unseen individuals, with CNN–LSTM achieving a mean accuracy of 97.21% ± 1.84% across 30 subjects. The results demonstrate that vision-based pose estimation with deep learning is a useful, precise, and non-intrusive approach to HAR in smart healthcare and home automation systems. date: 2026-02 publication: Sensors volume: 26 number: 5 pagerange: 1516 id_number: doi:10.3390/s26051516 refereed: TRUE issn: 1424-8220 official_url: http://doi.org/10.3390/s26051516 access: open language: en citation: Artículo Materias > Ingeniería Universidad Europea del Atlántico > Investigación > Producción Científica Fundación Universitaria Internacional de Colombia > Investigación > Producción Científica Universidad Internacional Iberoamericana México > Investigación > Artículos y libros Universidad Internacional Iberoamericana Puerto Rico > Investigación > Producción Científica Universidad Internacional do Cuanza > Investigación > Producción Científica Universidad de La Romana > Investigación > Producción Científica Abierto Inglés Human activity recognition (HAR) is essential in many applications, such as smart homes, assisted living, healthcare monitoring, rehabilitation, physiotherapy, and geriatric care. Conventional methods of HAR use wearable sensors, e.g., acceleration sensors and gyroscopes. However, they are limited by issues such as sensitivity to position, user inconvenience, and potential health risks with long-term use. Optical camera systems that are vision-based provide an alternative that is not intrusive; however, they are susceptible to variations in lighting, intrusions, and privacy issues. The paper uses an optical method of recognizing human domestic activities based on pose estimation and deep learning ensemble models. The skeletal keypoint features proposed in the current methodology are extracted from video data using PoseNet to generate a privacy-preserving representation that captures key motion dynamics without being sensitive to changes in appearance. A total of 30 subjects (15 male and 15 female) were sampled across 2734 activity samples, including nine daily domestic activities. There were six deep learning architectures, namely, the Transformer (Transformer), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Multilayer Perceptron (MLP), One-Dimensional Convolutional Neural Network (1D CNN), and a hybrid Convolutional Neural Network–Long Short-Term Memory (CNN–LSTM) architecture. The results on the hold-out test set show that the CNN–LSTM architecture achieves an accuracy of 98.78% within our experimental setting. Leave-One-Subject-Out cross-validation further confirms robust generalization across unseen individuals, with CNN–LSTM achieving a mean accuracy of 97.21% ± 1.84% across 30 subjects. The results demonstrate that vision-based pose estimation with deep learning is a useful, precise, and non-intrusive approach to HAR in smart healthcare and home automation systems. metadata Raza, Muhammad Amjad; Mehmood, Nasir; Siddiqui, Hafeez Ur Rehman; Saleem, Adil Ali; Álvarez, Roberto Marcelo; Miró Vera, Yini Airet y Díez, Isabel de la Torre mail SIN ESPECIFICAR, SIN ESPECIFICAR, SIN ESPECIFICAR, SIN ESPECIFICAR, roberto.alvarez@uneatlantico.es, yini.miro@uneatlantico.es, SIN ESPECIFICAR (2026) Human Activity Recognition in Domestic Settings Based on Optical Techniques and Ensemble Models. Sensors, 26 (5). p. 1516. ISSN 1424-8220 document_url: http://repositorio.unini.edu.mx/id/eprint/27968/1/sensors-26-01516-v2.pdf