☆ 4.7 Article

A WiFi-Based Method for Recognizing Fine-Grained Multiple-Subject Human Activities

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT (2023)

相关参考文献

注意：仅列出部分参考文献，下载原文获取全部文献信息。

Article Computer Science, Information Systems

RF-Based Human Activity Recognition Using Signal Adapted Convolutional Neural Network

Zhe Chen et al.

Summary: In this paper, a novel RF-based HAR framework called HAR-SAnet is introduced. By adaptively fusing features from both time and frequency domains into a neural network model, HAR-SAnet achieves higher recognition accuracy compared to existing algorithms and systems. Techniques such as point-wise grouped convolution and depth-wise separable convolutions are applied to optimize the model scale and inference execution time.

IEEE TRANSACTIONS ON MOBILE COMPUTING (2023)

添加到收藏夹

Article Engineering, Electrical & Electronic

Enabling Lightweight Device-Free Wireless Sensing With Network Pruning and Quantization

Ye Sun et al.

Summary: This paper introduces a lightweight device-free wireless sensing method based on edge intelligence techniques. By using network pruning and network quantization, the complexity and storage requirement of the sensing system can be reduced while maintaining good sensing performance.

IEEE SENSORS JOURNAL (2022)

添加到收藏夹

Article Computer Science, Information Systems

Push the Limit of Acoustic Gesture Recognition

Yanwen Wang et al.

Summary: This study proposes a robust contact-free gesture recognition system, RobuCIR, that utilizes frequency-hopping mechanism and data augmentation techniques to adapt to different practical impact factors. The results show that RobuCIR outperforms state-of-the-art works in terms of accuracy and robustness.

IEEE TRANSACTIONS ON MOBILE COMPUTING (2022)

添加到收藏夹

Article Chemistry, Analytical

HHI-AttentionNet: An Enhanced Human-Human Interaction Recognition Method Based on a Lightweight Deep Learning Model with Attention Network from CSI

Islam Md Shafiqul et al.

Summary: WiFi-HAR for human-human interaction recognition is a popular research topic, but the existing deep learning based methods often suffer from limited accuracy and high computational cost. In this study, a lightweight DL model, HHI-AttentionNet, is proposed to address these issues. The model achieves higher recognition accuracy by reducing the number of parameters and leveraging an antenna-frame-subcarrier attention mechanism. Experimental results demonstrate that HHI-AttentionNet outperforms existing methods with an average accuracy of 95.47% for recognizing 13 different HHIs.

SENSORS (2022)

添加到收藏夹

Article Computer Science, Information Systems

H2HI-Net: A Dual-Branch Network for Recognizing Human-to-Human Interactions From Channel-State Information

Mohamed Abdel-Basset et al.

Summary: This study addresses the challenge of human-to-human interaction recognition by proposing a novel device-free deep learning model named H2HI-NET, which models the Channel State Information of Wireless Fidelity devices for human activity recognition. Experimental evaluations demonstrate that H2HI-NET achieves higher accuracy compared to previous studies.

IEEE INTERNET OF THINGS JOURNAL (2022)

添加到收藏夹

Article Engineering, Electrical & Electronic

Edge Machine Learning Techniques Applied to RFID for Device-Free Hand Gesture Recognition

Massimo Merenda et al.

Summary: This research proposes a device-free gesture identification system that accurately recognizes different hand movements by processing the reflection signals of RFID tags. The use of Edge Machine Learning algorithms achieves a high accuracy rate of up to 99.4%.

IEEE JOURNAL OF RADIO FREQUENCY IDENTIFICATION (2022)

添加到收藏夹

Proceedings Paper Computer Science, Information Systems

An Approach for Recognizing Two-Human Interactions Using Channel State Information

Rami Alazrai et al.

Summary: A new CSI-based approach for recognizing two-human interactions was proposed in this paper. The approach consists of three phases: reducing the dimensionality of CSI data using principle component analysis, extracting features from the computed principal components, and building support vector machine classifiers using the selected features. Experimental results on a publicly available dataset showed an average accuracy of 86.21%, demonstrating the efficacy of the proposed approach in accurately recognizing various two-human interactions.

2022 9TH INTERNATIONAL CONFERENCE ON ELECTRICAL AND ELECTRONICS ENGINEERING (ICEEE 2022) (2022)

添加到收藏夹

Proceedings Paper Instruments & Instrumentation

A WiFi-based System for Recognizing Fine-grained Multiple-Subject Human Activities

Majid Ghosian Moghaddam et al.

Summary: This paper proposes a new approach for recognizing fine-grained human activities using CSI and RSSI WiFi data, achieving a high accuracy rate of 97.5%.

2022 IEEE INTERNATIONAL INSTRUMENTATION AND MEASUREMENT TECHNOLOGY CONFERENCE (I2MTC 2022) (2022)

添加到收藏夹

Proceedings Paper Instruments & Instrumentation

InARMS: Individual Activity Recognition of Multiple Subjects with FMCW radar

Hossein Raeis et al.

Summary: This paper proposes a non-learning approach for recognizing human activities in multi-subject environments using microwave Frequency-Modulated Continuous Wave (FMCW) radar. By applying mathematical and morphological operations on the range-Doppler map, the system is able to recognize activities in real time and measure the distance of targets.

2022 IEEE INTERNATIONAL INSTRUMENTATION AND MEASUREMENT TECHNOLOGY CONFERENCE (I2MTC 2022) (2022)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Activity recognition in a smart home using local feature weighting and variants of nearest-neighbors classifiers

Labiba Gillani Fahad et al.

Summary: Recognition of activities in a smart home can promote independent living of elderly individuals and extend their period of independent living. By assigning weights to features based on their importance in both inter-class and intra-class contexts, different activities can be accurately recognized.

JOURNAL OF AMBIENT INTELLIGENCE AND HUMANIZED COMPUTING (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Activity recognition and anomaly detection in smart homes

Labiba Gillani Fahad et al.

Summary: The proposed approach utilizes probabilistic neural network and H2O autoencoder for activity recognition and anomaly detection in smart homes. A comprehensive evaluation on two CASAS smart home datasets demonstrates its effectiveness in activity recognition and anomaly identification.

NEUROCOMPUTING (2021)

添加到收藏夹

Article Telecommunications

Multi-person device-free gesture recognition using mmWave signals

Jie Wang et al.

China Communications (2021)

添加到收藏夹

Article Chemistry, Analytical

A Domain-Independent Generative Adversarial Network for Activity Recognition Using WiFi CSI Data

Augustinas Zinys et al.

Summary: Research has shown the wide significance of device-free sensing using WiFi CSI data in recognizing human activities and gestures in various contexts. However, the performance of these techniques may degrade due to various factors, resulting in reduced accuracy. This paper proposes a domain-independent generative adversarial network in combination with a simplified data pre-processing module, which outperforms existing techniques in terms of robustness against domain change, accuracy of activity recognition, and model complexity.

SENSORS (2021)

添加到收藏夹

Article Computer Science, Information Systems

Multilocation Human Activity Recognition via MIMO-OFDM-Based Wireless Networks: An IoT-Inspired Device-Free Sensing Approach

Yi Zhong et al.

Summary: Device-free sensing (DFS) is an emerging technology that transforms traditional wireless communication networks into intelligent context-aware networks through machine-learning, opening doors for promising IoT applications. To address multi-location activity sensing challenges, an activity decomposition network (ActNet) is proposed to assemble data from different locations for training and achieve an average classification accuracy of 94.6%.

IEEE INTERNET OF THINGS JOURNAL (2021)

添加到收藏夹

Article Chemistry, Analytical

Wi-Fi-Based Location-Independent Human Activity Recognition via Meta Learning

Xue Ding et al.

Summary: The study introduces a Wi-Fi-based location-independent human activity recognition system, achieving accurate recognition with limited dataset. By utilizing methods such as Convolutional Neural Network and Long Short-Term Memory for feature representation, as well as metric learning-based activity recognition, it demonstrates good generalization ability and transferability.

SENSORS (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic

Human Activity Recognition with Device-Free Sensors for Well-Being Assessment in Smart Homes

Hossein Raeis et al.

IEEE INSTRUMENTATION & MEASUREMENT MAGAZINE (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic

RF-Wri: An Efficient Framework for RF-Based Device-Free Air-Writing Recognition

Can Uysal et al.

Summary: In the future, device-free technologies such as RF waves will be more commonly used for human-machine interactions, requiring machine learning and polarization diversity to improve accuracy. New feature extraction methods and signal processing techniques can make device-free air-writing recognition more effective.

IEEE SENSORS JOURNAL (2021)

添加到收藏夹

Article Computer Science, Information Systems

Multi-Adversarial In-Car Activity Recognition Using RFIDs

Fangxin Wang et al.

Summary: In-car human activity recognition presents an opportunity for intelligent driving behavior detection and touchless human-car interaction. While RFID technology offers advantages for activity recognition, existing solutions are limited to stable indoor spaces, making it challenging to adapt to the dynamic and compact interior of cars. Researchers have developed RF-CAR, a novel RFID-based solution that effectively filters domain-specific features in RF signals for robust activity recognition in diverse in-car environments. The solution combines deep learning and domain adaptation to achieve a high recognition accuracy of around 95 percent, surpassing the state-of-the-art solutions.

IEEE TRANSACTIONS ON MOBILE COMPUTING (2021)

添加到收藏夹

Article Chemistry, Analytical

Device-Free Human Activity Recognition with Low-Resolution Infrared Array Sensor Using Long Short-Term Memory Neural Network

Cunyi Yin et al.

Summary: This paper proposes a sensor-based human activity recognition method using a low-resolution infrared array sensor and deep learning framework, achieving a recognition accuracy of 98.287%. Compared to existing machine learning methods, the proposed approach yields better results and provides a low-cost yet promising solution for privacy-preserving scenarios.

SENSORS (2021)

添加到收藏夹

Proceedings Paper Computer Science, Artificial Intelligence

Activity & Emotion Detection of Recognized kids in CCTV Video for Day Care Using SlowFast & CNN

Narayana Darapaneni et al.

Summary: The project aims to process CCTV videos to identify children and their expressions, monitoring their activities at daycare. By tracking children's expressions and various activities, as well as creating a summary video of their activities, it enhances the overall experience for parents at the daycare.

2021 IEEE WORLD AI IOT CONGRESS (AIIOT) (2021)

添加到收藏夹

Proceedings Paper Automation & Control Systems

WiDG: An Air Hand Gesture Recognition System Based on CSI and Deep Learning

Zhengjie Wang et al.

Summary: In this paper, a device-free hand gesture recognition system based on CSI and deep learning models called WiDG is proposed, which can recognize handwritten digits from 0 to 9 in the air. Experimental results show that deep learning models can achieve satisfactory recognition accuracy using CSI.

PROCEEDINGS OF THE 33RD CHINESE CONTROL AND DECISION CONFERENCE (CCDC 2021) (2021)

添加到收藏夹

Proceedings Paper Acoustics

WIFI-BASED DEVICE-FREE GESTURE RECOGNITION THROUGH-THE-WALL

Sai Deepika Regani et al.

Summary: This work proposes a gesture recognition system that can accommodate an expandable set of gestures by leveraging the rich multipath in a through-the-wall setting. A statistical model for channel variations induced by hand gestures is developed, which is used to derive a correspondence between hand movement distance and Time Reversal Resonating Strength (TRRS) decay. Feature extraction modules are designed based on this relationship and the geometry of the gesture shape to enable gesture classification, achieving an 87% classification accuracy on a set of 6 uppercase English alphabets with a prototype built on off-the-shelf WIFI devices.

2021 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP 2021) (2021)

添加到收藏夹

Proceedings Paper Computer Science, Information Systems

Device-Free Human Activity Recognition With Identity-Based Transfer Mechanism

Bo Wu et al.

Summary: This paper introduces a human activity recognition system CrossID based on identity transfer mechanism, which addresses the major issue of device-free human activity recognition by using high-level personal characteristics as identities for training and transferring. Experimental results show that the system achieves a recognition rate of 95% in the target domain, demonstrating its effectiveness.

2021 IEEE WIRELESS COMMUNICATIONS AND NETWORKING CONFERENCE (WCNC) (2021)

添加到收藏夹

Proceedings Paper Computer Science, Theory & Methods

Wi-Fi-based Approach for Recognizing Human-Human Interactions

Rami Alazrai et al.

Summary: This paper introduces a new approach for recognizing human-human interactions using Wi-Fi signals, which analyzes CSI data and extracts features to construct a multi-class support vector machine classifier. The proposed method achieved an average recognition accuracy of 69.78% for 13 interactions, demonstrating the feasibility of recognizing human-human interactions using the CSI metric of the Wi-Fi signals.

2021 12TH INTERNATIONAL CONFERENCE ON INFORMATION AND COMMUNICATION SYSTEMS (ICICS) (2021)

添加到收藏夹

Article Computer Science, Information Systems

A Unified Approach for Patient Activity Recognition in Healthcare Using Depth Camera

Muhammad Hameed Siddiqi et al.

Summary: The research utilizes a depth camera to address security and privacy issues, and proposes an unsupervised segmentation algorithm for accurate human body segmentation from video frames. Additionally, a new feature selection technique is introduced to extract and select the best features from the feature space.

IEEE ACCESS (2021)

添加到收藏夹

Article Computer Science, Information Systems

Pantomime: Mid-Air Gesture Recognition with Sparse Millimeter-Wave Radar Point Clouds

Sameera Palipana et al.

Summary: Pantomime is a novel mid-air gesture recognition system that uses millimeter-wave RF signals, achieving high accuracy and AUC. The system demonstrates robustness across different environments and conditions, making it suitable for industrial, vehicular, and smart home applications.

PROCEEDINGS OF THE ACM ON INTERACTIVE MOBILE WEARABLE AND UBIQUITOUS TECHNOLOGIES-IMWUT (2021)

添加到收藏夹

Article Computer Science, Information Systems

CSI-IANet: An Inception Attention Network for Human-Human Interaction Recognition Based on CSI Signal

M. Humayun Kabir et al.

Summary: This paper proposes a human-human interaction (HHI) classifier, CSI-IANet, that utilizes a modified inception CNN with a spatial-attention mechanism. Experimental results show that the model achieved an average accuracy of 91.30%, which is 5% higher than the existing best method.

IEEE ACCESS (2021)

添加到收藏夹

Article Automation & Control Systems

Practical Device-Free Gesture Recognition Using WiFi Signals Based on Metalearning

Xiaorui Ma et al.

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS (2020)

添加到收藏夹

Article Computer Science, Information Systems

CSI-HC: A WiFi-Based Indoor Complex Human Motion Recognition Method

Zhanjun Hao et al.

MOBILE INFORMATION SYSTEMS (2020)

添加到收藏夹

Review Chemistry, Analytical

Human Activity Sensing with Wireless Signals: A Survey

Jiao Liu et al.

SENSORS (2020)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Fine-grained occupant activity monitoring with Wi-Fi channel state information: Practical implementation of multiple receiver settings

Hoonyong Lee et al.

ADVANCED ENGINEERING INFORMATICS (2020)

添加到收藏夹

Review Computer Science, Hardware & Architecture

A review and categorization of techniques on device-free human activity recognition

Zawar Hussain et al.

JOURNAL OF NETWORK AND COMPUTER APPLICATIONS (2020)

添加到收藏夹

Article Engineering, Multidisciplinary

The Recognition of Teacher Behavior Based on Multimodal Information Fusion

Dongli Wu et al.

MATHEMATICAL PROBLEMS IN ENGINEERING (2020)

添加到收藏夹

Article Computer Science, Information Systems

Patient Monitoring by Abnormal Human Activity Recognition Based on CNN Architecture

Malik Ali Gul et al.

ELECTRONICS (2020)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Recognizing activities of daily living from UWB radars and deep learning

Julien Maitre et al.

EXPERT SYSTEMS WITH APPLICATIONS (2020)

添加到收藏夹

Article Chemistry, Analytical

Accelerometer-Based Human Activity Recognition for Patient Monitoring Using a Deep Neural Network

Esther Fridriksdottir et al.

SENSORS (2020)

添加到收藏夹

Article Multidisciplinary Sciences

A dataset for Wi-Fi-based human-to-human interaction recognition

Rami Alazrai et al.

DATA IN BRIEF (2020)

添加到收藏夹

Article Computer Science, Information Systems

WiFind: Driver Fatigue Detection with Fine-Grained Wi-Fi Signal Features

Weijia Jia et al.

IEEE TRANSACTIONS ON BIG DATA (2020)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Device free human activity and fall recognition using WiFi channel state information (CSI)

Neena Damodaran et al.

CCF TRANSACTIONS ON PERVASIVE COMPUTING AND INTERACTION (2020)

添加到收藏夹

Article Computer Science, Information Systems

Sensing Technology for Human Activity Recognition: A Comprehensive Survey

Biying Fu et al.

IEEE ACCESS (2020)

添加到收藏夹

Article Computer Science, Information Systems

An End-to-End Deep Learning Framework for Recognizing Human-to-Human Interactions Using Wi-Fi Signals

Rami Alazrai et al.

IEEE ACCESS (2020)

添加到收藏夹

Article Computer Science, Theory & Methods

WiFi Sensing with Channel State Information: A Survey

Yongsen Ma et al.

ACM COMPUTING SURVEYS (2019)

添加到收藏夹

Article Chemistry, Analytical

Higher Order Feature Extraction and Selection for Robust Human Gesture Recognition using CSI of COTS Wi-Fi Devices

Hasmath Farhana Thariq Ahmed et al.

SENSORS (2019)

添加到收藏夹

Article Chemistry, Analytical

MCSM-Wri: A Small-Scale Motion Recognition Method Using WiFi Based on Multi-Scale Convolutional Neural Network

Shiyuan Ma et al.

SENSORS (2019)

添加到收藏夹

Article Computer Science, Information Systems

A Ubiquitous WiFi-Based Fine-Grained Gesture Recognition System

Heba Abdelnasser et al.

IEEE TRANSACTIONS ON MOBILE COMPUTING (2019)

添加到收藏夹

Review Automation & Control Systems

A review of state-of-the-art techniques for abnormal human activity recognition

Chhavi Dhiman et al.

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE (2019)

添加到收藏夹

Article Computer Science, Information Systems

Making Sense of Doppler Effect for Multi-Modal Hand Motion Detection

Wenjie Ruan et al.

IEEE TRANSACTIONS ON MOBILE COMPUTING (2018)

添加到收藏夹

Article Engineering, Electrical & Electronic

Activity Recognition for Incomplete Spinal Cord Injury Subjects Using Hidden Markov Models

Pichleap Sok et al.

IEEE SENSORS JOURNAL (2018)

添加到收藏夹

Article Computer Science, Information Systems

Device-Free Occupant Activity Sensing Using WiFi-Enabled IoT Devices for Smart Homes

Jianfei Yang et al.

IEEE INTERNET OF THINGS JOURNAL (2018)

添加到收藏夹

Article Telecommunications

WiDriver: Driver Activity Recognition System Based on WiFi CSI

Shihong Duan et al.

INTERNATIONAL JOURNAL OF WIRELESS INFORMATION NETWORKS (2018)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Recognizing Fine-Grained and Composite Activities Using Hand-Centric Features and Script Data

Marcus Rohrbach et al.

INTERNATIONAL JOURNAL OF COMPUTER VISION (2016)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Toward Personalized Activity Recognition Systems With a Semipopulation Approach

Jin-Hyuk Hong et al.

IEEE TRANSACTIONS ON HUMAN-MACHINE SYSTEMS (2016)

添加到收藏夹

Article Computer Science, Software Engineering

Multimodal Wearable Sensing for Fine-Grained Activity Recognition in Healthcare

Debraj De et al.

IEEE INTERNET COMPUTING (2015)

添加到收藏夹

Proceedings Paper Computer Science, Artificial Intelligence

An Improved Model for Segmentation and Recognition of Fine-grained Activities with Application to Surgical Training Tasks

Colin Lea et al.

2015 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV) (2015)

添加到收藏夹

Article Computer Science, Information Systems

Design and implementation of a system for body posture recognition

Ali Asghar Nazari Shirehjini et al.

MULTIMEDIA TOOLS AND APPLICATIONS (2014)

添加到收藏夹

Article Computer Science, Hardware & Architecture

Towards Collaborative Group Activity Recognition Using Mobile Devices

Dawud Gordon et al.

MOBILE NETWORKS & APPLICATIONS (2013)

添加到收藏夹

© Peeref 2019-2024. All rights reserved.