Results 71 to 80 of about 3,076 (167)

Intelligent Supportive System for People with Profound Intellectual and Multiple Disabilities

Advanced Intelligent Systems, Volume 8, Issue 1, January 2026.
A holistic INSENSION system is developed—a novel intelligent decision support system leveraging state‐of‐the‐art noninvasive audio‐visual sensor technologies together with machine learning algorithms and expert knowledge, to detect and interpret behaviors and communications (nonverbal signals—NVSs) of people with PIMD in challenging real‐world ...
Gašper Slapničar, Michał Kosiedowski, Arkadiusz Radziuk, Erik Dovgan, Torsten Hammann, Meike Engelhardt, Christin Kupitz, Carmen Campomanes‐Alvarez, Dorota Janczewska, Peter Zentel, Mitja Luštrek   +10 more
wiley   +1 more source

A Study on the Digital Application of OpenPose Skeleton Tracking Technology in Sports Dance Teaching in Chinese Universities from a Modernization Perspective

International Journal of Academic Research in Progressive Education and Development
With the modernization of sports, the application of digital technology in university physical education has gained increasing attention. This study explores the effectiveness of OpenPose skeleton tracking technology in sports dance instruction using a ...
Y. Ge, B. Abdullah, Hazizi Abu Saad
semanticscholar   +1 more source

A Human–Robot Collaboration Control Framework for Supernumerary Robotic Limbs

Journal of Field Robotics, Volume 43, Issue 1, Page 34-48, January 2026.
ABSTRACT In this paper, a novel human–robot collaboration (HRC) framework is proposed to enhance the capability of Supernumerary Robotic Limbs (SRLs). The proposed framework consists of three key modules: human motion intention recognition, disturbance rejection control, and obstacle avoidance, all of which are crucial for SRLs in the process of HRC ...
Jing Luo, Xiangyu Zhou, Yifan Zhu, Yu Li, Chaoyi Zhang, Keao Wang, Zhaohong Mai, Chao Zeng   +7 more
wiley   +1 more source

Evaluating the Accuracy of a Vision‐Based Algorithm for Groundline Estimation in Trotting Horses Using Multiple Camera Angles

Veterinary Medicine and Science, Volume 12, Issue 1, January 2026.
A markerless, computer vision‐based algorithm (RealHorse) was evaluated for estimating the groundline in trotting horses from multiple camera angles, including handheld recordings, on a treadmill. The algorithm showed high agreement with a fixed reference, with a groundline angle absolute error of 0.45° and stride‐level Maxdiff and Mindiff absolute ...
Karsten Key, Katja Berg, Jakob Kirkegaard, Kristian Ringkjær Andresen, Sabrina Skov Hansen   +4 more
wiley   +1 more source

Research on Motion Correction Method Based on Improved OpenPose and Fuzzy Decision

In response to the issues of traditional sports correction systems, which rely on rigid angle thresholds for action assessment leading to ambiguous classification boundaries, high costs of personal training services, and uneven distribution of resources,
Keqiong Chen, Chuanle Nie, Litao Fu, Chen Chen, Maofa Zhen, Yu Xie   +5 more
semanticscholar   +1 more source

CLS‐Transformer for Home Service Robots: A Lightweight Action Recognition Model Based on Skeleton and Native Length Encoding

IET Image Processing, Volume 20, Issue 1, January/December 2026.
The paper presents a native length encoding approach alongside a human skeleton‐based action recognition model (named CLS‐Transformer) for home service robots. The model incorporates a novel cue coding strategy that captures richer feature information. The experimental results demonstrate that CLS‐Transformer achieves superior recognition accuracy and ...
Jie Shen, Yuming Zhang, Xiaoyang Li, Meng Zhou, Wenfan Wang, Dawei Zhang   +5 more
wiley   +1 more source

Research on Lightweight Traffic Police Gesture Recognition Based on Optical Flow Enhancement

IET Intelligent Transport Systems, Volume 20, Issue 1, January/December 2026.
This study proposes a two‐stage optical flow‐enhanced framework for traffic police gesture recognition. The first stage employs the RAFT algorithm to extract precise motion vectors while suppressing background interference, followed by a lightweight ShuffleNet‐K5 network with 5 × 5 depthwise separable convolution and a novel Simplified State Attention (
Mengying Chang, Huizhi Xu, Yongshuai Ge
wiley   +1 more source

GSDF‐Gait: A GCN and Self‐Attention Dynamic Fusion Network for Gait‐Based Person Recognition

Applied Computational Intelligence and Soft Computing, Volume 2026, Issue 1, 2026.
Graph convolutional networks (GCNs) are extensively used for skeleton‐based gait recognition. Nevertheless, despite significant improvements, a substantial challenge lies in the restricted receptive field of GCNs. However, separate structural joints could also reveal a notably important correlation.
Md. Khaliluzzaman, Kaushik Deb, Nasimul Noman   +2 more
wiley   +1 more source

OpenBeePose: A Remote Sensing Framework for Bee Pose Estimation Using Deep Learning

Applied Sciences
The application of remote sensing technology for pose estimation in beekeeping has the potential to transform colony management, improve bee health and mitigate the decline in bee populations.
Sajad Sabzi, Ali Najar, Raziyeh Pourdarbani, Ginés García-Mateos, Ruben Fernandez-Beltran, Mohammad H. Rohban   +5 more
doaj   +1 more source

Comparative Analysis of Lightweight OpenPose and MoveNet AI Models for Real-time Fall Detection and Alert Systems

Sensors and materials
.
Yuh-Shihng Chang, Yi-Xiang Zheng, Zheng-Yu Ku   +2 more
semanticscholar   +1 more source