Face hallucination with application in far distance face recognition

In this thesis, faces captured in far distances are investigated. Face enhancement algorithms are studied. Hallucinating faces in holistic model and patch-based model are analysed respectively. The advantages and disadvantages of both models are discussed. An innovative holistic model and patch-base...

Full description

Bibliographic Details
Main Author: Xu, Xiang
Format: Thesis
Language:English
Published: Curtin University 2014
Online Access:http://hdl.handle.net/20.500.11937/1931
Description
Summary:In this thesis, faces captured in far distances are investigated. Face enhancement algorithms are studied. Hallucinating faces in holistic model and patch-based model are analysed respectively. The advantages and disadvantages of both models are discussed. An innovative holistic model and patch-based model are proposed separately. More investigation in practical surveillance environments are carried. And a new far face recognition model is proposed. Experiments demonstrate the improvement of proposed approaches.