Intelligent Video Analytics for Human Action Recognition: The State of Knowledge

Marek Kulbacki; Jakub Segen; Zenon Chaczko; Jerzy W. Rozenblit; Michał Kulbacki; Ryszard Klempous; Konrad Wojciechowski

doi:10.3390/s23094258

Intelligent Video Analytics for Human Action Recognition: The State of Knowledge

Marek Kulbacki, Jakub Segen, Zenon Chaczko, Jerzy W. Rozenblit, Michał Kulbacki, Ryszard Klempous, Konrad Wojciechowski

Research output: Contribution to journal › Review article › peer-review

4 Scopus citations

Abstract

The paper presents a comprehensive overview of intelligent video analytics and human action recognition methods. The article provides an overview of the current state of knowledge in the field of human activity recognition, including various techniques such as pose-based, tracking-based, spatio-temporal, and deep learning-based approaches, including visual transformers. We also discuss the challenges and limitations of these techniques and the potential of modern edge AI architectures to enable real-time human action recognition in resource-constrained environments.

Original language	English (US)
Article number	4258
Journal	Sensors
Volume	23
Issue number	9
DOIs	https://doi.org/10.3390/s23094258
State	Published - May 2023

Keywords

deep learning-based HAR
edge AI
human activity recognition
intelligent video analytics
pose-based HAR
spatio-temporal-based HAR
tracking-based HAR
video surveillance
visual transformers

ASJC Scopus subject areas

Analytical Chemistry
Information Systems
Atomic and Molecular Physics, and Optics
Biochemistry
Instrumentation
Electrical and Electronic Engineering

Access to Document

10.3390/s23094258

Cite this

@article{9da8a46416a9434fab7675f9bbcc5062,

title = "Intelligent Video Analytics for Human Action Recognition: The State of Knowledge",

abstract = "The paper presents a comprehensive overview of intelligent video analytics and human action recognition methods. The article provides an overview of the current state of knowledge in the field of human activity recognition, including various techniques such as pose-based, tracking-based, spatio-temporal, and deep learning-based approaches, including visual transformers. We also discuss the challenges and limitations of these techniques and the potential of modern edge AI architectures to enable real-time human action recognition in resource-constrained environments.",

keywords = "deep learning-based HAR, edge AI, human activity recognition, intelligent video analytics, pose-based HAR, spatio-temporal-based HAR, tracking-based HAR, video surveillance, visual transformers",

author = "Marek Kulbacki and Jakub Segen and Zenon Chaczko and Rozenblit, {Jerzy W.} and Micha{\l} Kulbacki and Ryszard Klempous and Konrad Wojciechowski",

note = "Publisher Copyright: {\textcopyright} 2023 by the authors.",

year = "2023",

month = may,

doi = "10.3390/s23094258",

language = "English (US)",

volume = "23",

journal = "Sensors",

issn = "1424-8220",

publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",

number = "9",

}

TY - JOUR

T1 - Intelligent Video Analytics for Human Action Recognition

T2 - The State of Knowledge

AU - Kulbacki, Marek

AU - Segen, Jakub

AU - Chaczko, Zenon

AU - Rozenblit, Jerzy W.

AU - Kulbacki, Michał

AU - Klempous, Ryszard

AU - Wojciechowski, Konrad

PY - 2023/5

Y1 - 2023/5

N2 - The paper presents a comprehensive overview of intelligent video analytics and human action recognition methods. The article provides an overview of the current state of knowledge in the field of human activity recognition, including various techniques such as pose-based, tracking-based, spatio-temporal, and deep learning-based approaches, including visual transformers. We also discuss the challenges and limitations of these techniques and the potential of modern edge AI architectures to enable real-time human action recognition in resource-constrained environments.

AB - The paper presents a comprehensive overview of intelligent video analytics and human action recognition methods. The article provides an overview of the current state of knowledge in the field of human activity recognition, including various techniques such as pose-based, tracking-based, spatio-temporal, and deep learning-based approaches, including visual transformers. We also discuss the challenges and limitations of these techniques and the potential of modern edge AI architectures to enable real-time human action recognition in resource-constrained environments.

KW - deep learning-based HAR

KW - edge AI

KW - human activity recognition

KW - intelligent video analytics

KW - pose-based HAR

KW - spatio-temporal-based HAR

KW - tracking-based HAR

KW - video surveillance

KW - visual transformers

UR - http://www.scopus.com/inward/record.url?scp=85159268819&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85159268819&partnerID=8YFLogxK

U2 - 10.3390/s23094258

DO - 10.3390/s23094258

M3 - Review article

C2 - 37177461

AN - SCOPUS:85159268819

SN - 1424-8220

VL - 23

JO - Sensors

JF - Sensors

IS - 9

M1 - 4258

ER -

Intelligent Video Analytics for Human Action Recognition: The State of Knowledge

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this