Anglia Ruskin Research Online (ARRO)
Browse
Constantinou_2018.pdf (2.97 MB)

Use of a high resolution 3D optical scanner for 3D model creation, game design and facial expression recognition

Download (2.97 MB)
thesis
posted on 2023-08-30, 15:48 authored by Georgia Constantinou
The process of three-dimensional (3D) scanning uses various techniques to capture the shape of an object in a computer file using a 3D scanner. This current research utilises a new camera-based 3D scanning technology (Mephisto Extreme 3D Optical scanner) that can very rapidly acquire high resolution 3D object models. The aims of this work include the configuration, assessment and evaluation of this 3D scanner to optimise scan quality, improve 3D object processing techniques that integrate 3D scanning, model construction and computer game development and evaluation of the use of the scanner for acquisition of facial features and its potential use in facial expression recognition. A procedure is presented detailing the configuration settings that will maximise 3D scan quality. The successful acquisition of numerous high quality 3D models from a variety of small inanimate and face target sources is reported. Appropriate graphics modelling software that can process 3D objects from acquisition and/or creation (Mephisto Extreme, 3DS Max) through to game import (Unity 3D) is presented and highlights the importance of facilitating portability of the 3D object models in the process chain. An OBJ file format reader/writer is developed where proof-of-principle is established that object model data output from the scanner can be easily and quickly extracted and potentially processed prior to input into a suitable game engine. This scanning-processing technique could potentially reduce the game design and development time from months/weeks to a few days. Other results include the successful scans of 3D facial expressions, and some possibilities for how this work could further progress research in 2D facial expression recognition are explored.

History

Institution

Anglia Ruskin University

File version

  • Accepted version

Language

  • eng

Thesis name

  • MPhil

Thesis type

  • Masters

Legacy posted date

2018-11-07

Legacy creation date

2018-11-07

Legacy Faculty/School/Department

Theses from Anglia Ruskin University/Faculty of Science and Technology

Usage metrics

    ARU Theses

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC