<> "The repository administrator has not yet configured an RDF license."^^ . <> . . . "Human Activity Recognition in Domestic Settings Based on Optical Techniques and Ensemble Models"^^ . "Human activity recognition (HAR) is essential in many applications, such as smart homes, assisted\n living, healthcare monitoring, rehabilitation, physiotherapy, and geriatric care. Conventional methods of\n HAR use wearable sensors, e.g., acceleration sensors and gyroscopes. However, they are limited by issues\n such as sensitivity to position, user inconvenience, and potential health risks with long-term use. Optical\n camera systems that are vision-based provide an alternative that is not intrusive; however, they are\n susceptible to variations in lighting, intrusions, and privacy issues. The paper uses an optical method of\n recognizing human domestic activities based on pose estimation and deep learning ensemble models. The\n skeletal keypoint features proposed in the current methodology are extracted from video data using PoseNet\n to generate a privacy-preserving representation that captures key motion dynamics without being sensitive to\n changes in appearance. A total of 30 subjects (15 male and 15 female) were sampled across 2734 activity\n samples, including nine daily domestic activities. There were six deep learning architectures, namely, the\n Transformer (Transformer), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Multilayer Perceptron\n (MLP), One-Dimensional Convolutional Neural Network (1D CNN), and a hybrid Convolutional Neural Network–Long\n Short-Term Memory (CNN–LSTM) architecture. The results on the hold-out test set show that the CNN–LSTM\n architecture achieves an accuracy of 98.78% within our experimental setting. Leave-One-Subject-Out\n cross-validation further confirms robust generalization across unseen individuals, with CNN–LSTM achieving a\n mean accuracy of 97.21% ± 1.84% across 30 subjects. The results demonstrate that vision-based pose\n estimation with deep learning is a useful, precise, and non-intrusive approach to HAR in smart healthcare\n and home automation systems."^^ . "2026-02" . . . "26" . "5" . . "Sensors"^^ . . . "14248220" . . . . . . . . . . . . . . . . . . . . . . . . . "Hafeez Ur Rehman"^^ . "Siddiqui"^^ . "Hafeez Ur Rehman Siddiqui"^^ . . "Roberto Marcelo"^^ . "Álvarez"^^ . "Roberto Marcelo Álvarez"^^ . . "Muhammad Amjad"^^ . "Raza"^^ . "Muhammad Amjad Raza"^^ . . "Isabel de la Torre"^^ . "Díez"^^ . "Isabel de la Torre Díez"^^ . . "Yini Airet"^^ . "Miró Vera"^^ . "Yini Airet Miró Vera"^^ . . "Nasir"^^ . "Mehmood"^^ . "Nasir Mehmood"^^ . . "Adil Ali"^^ . "Saleem"^^ . "Adil Ali Saleem"^^ . . . . . . "Human Activity Recognition in Domestic Settings Based on Optical Techniques and Ensemble Models (Texto)"^^ . . . "sensors-26-01516-v2.pdf"^^ . . . "Human Activity Recognition in Domestic Settings Based on Optical Techniques and Ensemble Models (Otro)"^^ . . . . . . "indexcodes.txt"^^ . . "HTML Summary of #27968 \n\nHuman Activity Recognition in Domestic Settings Based on Optical Techniques and Ensemble Models\n\n" . "text/html" . . . "Engineering"@en . "Ingeniería"@es . .