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 |
|
| Supervisors/Advisors |
|
| Thesis sponsors | |
| Award date | 26 Jan 2020 |
| DOIs | |
| Publication status | Unpublished - 2020 |
Access to Document
- 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.