Data-driven robust human action recognition

Jian Liu

doi:10.26182/5e4dd397f16a9

Data-driven robust human action recognition

Jian Liu

School of Physics, Maths and Computing

Research output: Thesis › Doctoral Thesis

61 Downloads (Pure)

Abstract

Human action recognition has wide applications in security, entertainment, sports, and medicine. It is very challenging to develop robust action recognition models for real-life scenarios, where human movements are associated with numerous irrelevant variations such as viewpoints, backgrounds, and illuminations. This thesis proposes novel data-driven methods to address the fundamental problems in robust action recognition, including human data synthesis, spatio-temporal representation, action classification, and human pose recovery. Besides, the adversarial robustness of human action recognition is investigated in this thesis.

Original language	English
Qualification	Doctor of Philosophy
Awarding Institution	The University of Western Australia
Thesis sponsors	Australian Government Research Training Program
Award date	26 Jan 2020
DOIs	https://doi.org/10.26182/5e4dd397f16a9
Publication status	Unpublished - 2020

Fingerprint

Cite this

Liu, J. (2020). Data-driven robust human action recognition. https://doi.org/10.26182/5e4dd397f16a9

@phdthesis{0621a6c510d84745b6397b049ebf480b,

title = "Data-driven robust human action recognition",

abstract = "Human action recognition has wide applications in security, entertainment, sports, and medicine. It is very challenging to develop robust action recognition models for real-life scenarios, where human movements are associated with numerous irrelevant variations such as viewpoints, backgrounds, and illuminations. This thesis proposes novel data-driven methods to address the fundamental problems in robust action recognition, including human data synthesis, spatio-temporal representation, action classification, and human pose recovery. Besides, the adversarial robustness of human action recognition is investigated in this thesis.",

keywords = "Computer vision, data synthesis, deep learning, adversarial learning, Action Recognition, Pose recovery",

author = "Jian Liu",

year = "2020",

doi = "10.26182/5e4dd397f16a9",

language = "English",

school = "The University of Western Australia",

}

TY - THES

T1 - Data-driven robust human action recognition

AU - Liu, Jian

PY - 2020

Y1 - 2020

N2 - Human action recognition has wide applications in security, entertainment, sports, and medicine. It is very challenging to develop robust action recognition models for real-life scenarios, where human movements are associated with numerous irrelevant variations such as viewpoints, backgrounds, and illuminations. This thesis proposes novel data-driven methods to address the fundamental problems in robust action recognition, including human data synthesis, spatio-temporal representation, action classification, and human pose recovery. Besides, the adversarial robustness of human action recognition is investigated in this thesis.

AB - Human action recognition has wide applications in security, entertainment, sports, and medicine. It is very challenging to develop robust action recognition models for real-life scenarios, where human movements are associated with numerous irrelevant variations such as viewpoints, backgrounds, and illuminations. This thesis proposes novel data-driven methods to address the fundamental problems in robust action recognition, including human data synthesis, spatio-temporal representation, action classification, and human pose recovery. Besides, the adversarial robustness of human action recognition is investigated in this thesis.

KW - Computer vision

KW - data synthesis

KW - deep learning

KW - adversarial learning

KW - Action Recognition

KW - Pose recovery

U2 - 10.26182/5e4dd397f16a9

DO - 10.26182/5e4dd397f16a9

M3 - Doctoral Thesis

ER -

Access to Document

10.26182/5e4dd397f16a9

THESIS_DOCTOR_OF_PHILOSOPHY_LIU_Jian
This work is protected by Copyright. You may print or download ONE copy of this document for the purpose of your own non-commercial research or study. Any other use requires permission from the copyright owner. The Copyright Act requires you to attribute any copyright works you quote or paraphrase.