☆ 4.6 Article

Human Activity Recognition via Hybrid Deep Learning Based Model

SENSORS (2022)

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, Theory & Methods

Artificial Intelligence of Things-assisted two-stream neural network for anomaly detection in surveillance Big Video Data

Waseem Ullah et al.

Summary: This work presents an efficient and robust framework for recognizing anomalies in surveillance Big Video Data (BVD) using Artificial Intelligence of Things (AIoT). The framework consists of two phases: instant anomaly detection and detailed anomaly analysis. A two-stream neural network is proposed to model and classify the anomalies. Extensive experiments show that the framework outperforms state-of-the-art methods in terms of accuracy.

FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE (2022)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Symbiotic Graph Neural Networks for 3D Skeleton-Based Human Action Recognition and Motion Prediction

Maosen Li et al.

Summary: In this paper, we propose a symbiotic graph neural network model to address the problems of 3D skeleton-based action recognition and motion prediction. The model utilizes graphs of two scales to capture relations among body-joints and body-parts, and achieves better performances on both tasks through joint training of a backbone, an action-recognition head, and a motion-prediction head.

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE (2022)

添加到收藏夹

Article Computer Science, Artificial Intelligence

SD-Net: Understanding overcrowded scenes in real-time via an efficient dilated convolutional neural network

Noman Khan et al.

Summary: The advancements in computer vision technologies have attracted researchers to focus on surveillance applications, particularly in automated crowded scene analysis and crowd counting in real-time. In this study, a new end-to-end CNN architecture named SD-Net is proposed to generate high-quality density maps and accurately count people in extremely crowded scenes. The architecture utilizes depthwise separable, standard, and dilated 2D convolutional layers to address challenges such as cluttered backgrounds, occlusion, and limited training data coverage.

JOURNAL OF REAL-TIME IMAGE PROCESSING (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Efficient activity recognition using lightweight CNN and DS-GRU network for surveillance applications

Amin Ullah et al.

Summary: This paper proposes a lightweight deep learning-assisted framework for activity recognition, which detects and tracks humans in surveillance videos using CNN models, and learns temporal changes in frame sequences for activity recognition using DS-GRU. Experimental results demonstrate the efficiency of this technique for real-time surveillance applications.

APPLIED SOFT COMPUTING (2021)

添加到收藏夹

Article Computer Science, Information Systems

Deep-ReID: deep features and autoencoder assisted image patching strategy for person re-identification in smart cities surveillance

Samee Ullah Khan et al.

Summary: This study presents a unified framework incorporating learned features and deep autoencoder for Person Re-identification (P-ReID) task. It provides a trade-off between time complexity and accuracy, where a lightweight model can be employed with reduced number of autoencoder layers to achieve fast and flexible results.

MULTIMEDIA TOOLS AND APPLICATIONS (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Conflux LSTMs Network: A Novel Approach for Multi-View Action Recognition

Amin Ullah et al.

Summary: The paper introduces a conflux long short-term memory (LSTMs) network for action recognition from multi-view cameras. By utilizing four major steps, the framework successfully extracts features from different views for effective action recognition, resulting in performance improvement in experimental results.

NEUROCOMPUTING (2021)

添加到收藏夹

Article Computer Science, Theory & Methods

Human action recognition using attention based LSTM network with dilated CNN features

Khan Muhammad et al.

Summary: This study introduces a novel approach combining BiLSTM and DCNN with an attention mechanism to effectively recognize different human actions in videos, improving the performance of video-based action classification.

FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic

DB-Net: A novel dilated CNN based multi-step forecasting model for power consumption in integrated local energy systems

Noman Khan et al.

Summary: In the current era of cutting edge technology, the demand for electricity is increasing due to exponential population growth, electricity reliant vehicles, and home appliances. Precise energy consumption prediction and integrated local energy systems are crucial for clean energy management between consumers and suppliers. The hybrid network model 'DB-Net', incorporating DCNN and BiLSTM, improves efficiency in long- and short-term energy consumption prediction.

INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS (2021)

添加到收藏夹

Article Chemistry, Analytical

Atrous Convolutions and Residual GRU Based Architecture for Matching Power Demand with Supply

Samee Ullah Khan et al.

Summary: This study proposes an intelligent deep learning framework that integrates different neural network layers to match power demand with supply, accurately predict short-term energy demand, and provide effective communication methods. Through data acquisition, preprocessing, and feature extraction, the sequential learning model is utilized to optimize energy management, achieving better performance compared to existing approaches.

SENSORS (2021)

添加到收藏夹

Article Mathematics

Sequential Learning-Based Energy Consumption Prediction Model for Residential and Commercial Sectors

Ijaz Ul Haq et al.

Summary: The use of electrical energy is directly proportional to the increase in global population, driving researchers to develop forecasting models for optimal energy use. This paper presents a comparative analysis of deep features and sequential learning models to select an optimized hybrid architecture for energy consumption prediction, achieving superior results compared to existing models.

MATHEMATICS (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence

MLT-DNet: Speech emotion recognition using 1D dilated CNN based on multi-learning trick approach

Mustaqeem et al.

Summary: The paper discusses an end-to-end real-time SER model based on DCNN, which extracts emotional features and learns long-term contextual dependencies through multi-learning strategy, achieving high recognition accuracy.

EXPERT SYSTEMS WITH APPLICATIONS (2021)

添加到收藏夹

Article Mathematics

AB-Net: A Novel Deep Learning Assisted Framework for Renewable Energy Generation Forecasting

Noman Khan et al.

Summary: The article introduces a new architecture AB-Net for intelligent matching of renewable energy generation and consumption for efficient energy management. This method combines an autoencoder and bidirectional long short-term memory network in steps such as data acquisition, preprocessing, and feature extraction, achieving good prediction results.

MATHEMATICS (2021)

添加到收藏夹

Article Chemistry, Analytical

An Efficient Anomaly Recognition Framework Using an Attention Residual LSTM in Surveillance Videos

Waseem Ullah et al.

Summary: The study introduces an efficient lightweight anomaly recognition framework for smart city video surveillance, reducing time complexity and demonstrating superior performance in experiments.

SENSORS (2021)

添加到收藏夹

Article Mathematics

Diving Deep into Short-Term Electricity Load Forecasting: Comparative Analysis and a Novel Framework

Fath U. Min Ullah et al.

Summary: This study compares the performance of traditional and sequential learning algorithms in electricity load forecasting and presents a new ECP framework that utilizes deep sequential learning models to achieve optimal model optimization.

MATHEMATICS (2021)

添加到收藏夹

Article Computer Science, Information Systems

Deep Convolutional Neural Network with RNNs for Complex Activity Recognition Using Wrist-Worn Wearable Sensor Data

Sakorn Mekruksavanich et al.

Summary: This study focused on sensor-based human activity recognition using deep learning models, specifically convolutional and recurrent neural networks, to identify complex activities. By examining four RNN-based DL models and hybrid DL models combining convolutional layers, it was demonstrated that the CNNBiGRU model performed effectively in various scenarios, achieving high recognition accuracy.

ELECTRONICS (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Light-DehazeNet: A Novel Lightweight CNN Architecture for Single Image Dehazing

Hayat Ullah et al.

Summary: This study introduces a lightweight convolutional neural network, LD-Net, for reconstructing hazy images and proposes a color visibility restoration method to overcome weather challenges. Extensive experiments validate the superiority of the proposed method in image dehazing techniques and object detection tasks.

IEEE TRANSACTIONS ON IMAGE PROCESSING (2021)

添加到收藏夹

Article Computer Science, Information Systems

1D-CNN: Speech Emotion Recognition System Using a Stacked Network with Dilated CNN Features

Mustaqeem et al.

Summary: Researchers have developed new techniques in emotion recognition field, utilizing one-dimensional convolutional neural network and bidirectional gated recurrent unit to improve emotion recognition accuracy. Their model achieved different recognition rates on three benchmark datasets.

CMC-COMPUTERS MATERIALS & CONTINUA (2021)

添加到收藏夹

Article Computer Science, Information Systems

Batteries State of Health Estimation via Efficient Neural Networks With Multiple Channel Charging Profiles

Noman Khan et al.

Summary: PHM plays a main role in handling the risk of failure, accurate capacity estimation of batteries in ESS is mandatory for their safe operations, a MCCPs based BMS is proposed for estimating batteries health/capacity through deep learning.

IEEE ACCESS (2021)

添加到收藏夹

Article Chemistry, Analytical

Towards Efficient Electricity Forecasting in Residential and Commercial Buildings: A Novel Hybrid CNN with a LSTM-AE based Framework

Zulfiqar Ahmad Khan et al.

SENSORS (2020)

添加到收藏夹

Article Engineering, Chemical

MPPIF-Net: Identification of Plasmodium Falciparum Parasite Mitochondrial Proteins Using Deep Features with Multilayer Bi-directional LSTM

Samee Ullah Khan et al.

PROCESSES (2020)

添加到收藏夹

Article Mathematics

CLSTM: Deep Feature-Based Speech Emotion Recognition Using the Hierarchical ConvLSTM Network

Mustageem et al.

MATHEMATICS (2020)

添加到收藏夹

Article Chemistry, Analytical

Towards Efficient Building Designing: Heating and Cooling Load Prediction via Multi-Output Model

Muhammad Sajjad et al.

SENSORS (2020)

添加到收藏夹

Article Computer Science, Information Systems

Depth-Camera-Based System for Estimating Energy Expenditure of Physical Activities in Gyms

Bor-Shing Lin et al.

IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS (2019)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Human activity recognition using 2D skeleton data and supervised machine learning

Sumaira Ghazal et al.

IET IMAGE PROCESSING (2019)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Skeleton-based human activity recognition for elderly monitoring systems

Youssef Hbali et al.

IET COMPUTER VISION (2018)

添加到收藏夹

Article Chemistry, Analytical

Iss2Image: A Novel Signal-Encoding Technique for CNN-Based Human Activity Recognition

Taeho Hur et al.

SENSORS (2018)

添加到收藏夹

Article Computer Science, Software Engineering

Percentage of human-occupied areas for fall detection from two views

Mikael A. Mousse et al.

VISUAL COMPUTER (2017)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks

Shaoqing Ren et al.

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE (2017)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Automatic Learning of Articulated Skeletons Based on Mean of 3D Joints for Efficient Action Recognition

Abdelouahid Ben Tamou et al.

INTERNATIONAL JOURNAL OF PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE (2017)

添加到收藏夹

Article Chemistry, Analytical

A Human Activity Recognition System Based on Dynamic Clustering of Skeleton Data

Alessandro Manzi et al.

SENSORS (2017)

添加到收藏夹

Article Computer Science, Information Systems

Effective Active Skeleton Representation for Low Latency Human Action Recognition

Xingyang Cai et al.

IEEE TRANSACTIONS ON MULTIMEDIA (2016)

添加到收藏夹

Article Chemistry, Analytical

An Online Continuous Human Action Recognition Algorithm Based on the Kinect Sensor

Guangming Zhu et al.

SENSORS (2016)

添加到收藏夹

Article Computer Science, Information Systems

Sequence of the most informative joints (SMIJ): A new representation for human skeletal action recognition

Ferda Ofli et al.

JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION (2014)

添加到收藏夹

Article Geriatrics & Gerontology

Physical Activity Measurement in Older Adults: Relationships With Mental Health

Sarah J. Parker et al.

JOURNAL OF AGING AND PHYSICAL ACTIVITY (2008)

添加到收藏夹

© Peeref 2019-2024. All rights reserved.