RGB-D based human action recognition: From handcrafted to deep learning

Bangli Liu; Honghai Liu

doi:10.1002/9781119863663.ch26

RGB-D based human action recognition: From handcrafted to deep learning

Bangli Liu^*
, Honghai Liu

^*Corresponding author for this work

De Montfort University
University of Portsmouth

Research output: Chapter in Book/Report/Conference proceeding › Chapter › peer-review

Abstract

RGB-D based human action recognition has been a hot research topic with the release of RGB-D devices. Many attempts have been done to achieve robust and effective action recognition. This chapter reviews human action recognition techniques, including handcrafted feature representations extracted from different data modality and various deep neural network architectures. Moreover, commonly used action datasets, performance comparison, and promising future directions are presented.

Original language	English
Title of host publication	Handbook of Human-Machine Systems
Publisher	wiley
Pages	307-320
Number of pages	14
ISBN (Electronic)	9781119863663
ISBN (Print)	9781119863632
DOIs	https://doi.org/10.1002/9781119863663.ch26
State	Published - 7 Jul 2023
Externally published	Yes

Keywords

Action recognition
Deep learning
Handcrafted
RGB-D data
Review

Access to Document

10.1002/9781119863663.ch26

Cite this

@inbook{e21d1004348848d9824102ff5220dc7a,

title = "RGB-D based human action recognition: From handcrafted to deep learning",

abstract = "RGB-D based human action recognition has been a hot research topic with the release of RGB-D devices. Many attempts have been done to achieve robust and effective action recognition. This chapter reviews human action recognition techniques, including handcrafted feature representations extracted from different data modality and various deep neural network architectures. Moreover, commonly used action datasets, performance comparison, and promising future directions are presented.",

keywords = "Action recognition, Deep learning, Handcrafted, RGB-D data, Review",

author = "Bangli Liu and Honghai Liu",

year = "2023",

month = jul,

day = "7",

doi = "10.1002/9781119863663.ch26",

language = "英语",

isbn = "9781119863632",

pages = "307--320",

booktitle = "Handbook of Human-Machine Systems",

publisher = "wiley",

}

TY - CHAP

T1 - RGB-D based human action recognition

T2 - From handcrafted to deep learning

AU - Liu, Bangli

AU - Liu, Honghai

PY - 2023/7/7

Y1 - 2023/7/7

N2 - RGB-D based human action recognition has been a hot research topic with the release of RGB-D devices. Many attempts have been done to achieve robust and effective action recognition. This chapter reviews human action recognition techniques, including handcrafted feature representations extracted from different data modality and various deep neural network architectures. Moreover, commonly used action datasets, performance comparison, and promising future directions are presented.

AB - RGB-D based human action recognition has been a hot research topic with the release of RGB-D devices. Many attempts have been done to achieve robust and effective action recognition. This chapter reviews human action recognition techniques, including handcrafted feature representations extracted from different data modality and various deep neural network architectures. Moreover, commonly used action datasets, performance comparison, and promising future directions are presented.

KW - Action recognition

KW - Deep learning

KW - Handcrafted

KW - RGB-D data

KW - Review

UR - https://www.scopus.com/pages/publications/85168509281

U2 - 10.1002/9781119863663.ch26

DO - 10.1002/9781119863663.ch26

M3 - 章节

AN - SCOPUS:85168509281

SN - 9781119863632

SP - 307

EP - 320

BT - Handbook of Human-Machine Systems

PB - wiley

ER -

RGB-D based human action recognition: From handcrafted to deep learning

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this