Monash University
Browse

Towards Autonomous Training of Scene-Specific Pedestrian Detectors in Visual Surveillance Environments

Download (12.62 MB)
Version 2 2019-03-26, 01:50
Version 1 2019-03-25, 02:55
thesis
posted on 2019-03-26, 01:50 authored by MOHAMMED SHAH NEWAZ CHOWDHURY
Pedestrian detection is of paramount importance for intelligent visual surveillance of people. Despite over two decades of extensive research, the performance of generic pedestrian detectors remains prone to the dataset shift phenomenon, whereby target surveillance scenes differ significantly from the training data used to generate the detector. Numerous scene-specific approaches have been developed, but they are limited by their requirement for manual labeling or dependence on prior models. This research focuses on the development of a virtually autonomous training (VAT) framework that trains scene-specific pedestrian detectors without requiring any manual labeling, nor utilizing any prior model or training data.

History

Campus location

Malaysia

Principal supervisor

Poh Phaik Eong

Additional supervisor 1

Kuang Ye Chow

Year of Award

2019

Department, School or Centre

School of Engineering (Monash University Malaysia)

Course

Doctor of Philosophy

Degree Type

DOCTORATE

Faculty

Faculty of Engineering

Usage metrics

    Faculty of Engineering Theses

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC