Leeds Beckett University
Browse

A Parametric Framework and Tool for Posing Photorealistic Facial Expressions in 3D Characters

Download (9.54 MB)
thesis
posted on 2025-04-01, 16:14 authored by Amrinder RomanaAmrinder Romana

This research introduces a parametric framework to address the lack of educational resources for teaching realistic facial animation in digital characters, focusing on developing a Parametric Framework and Tool for Posing Photorealistic 3D Character Facial Expressions. The research applies the Facial Action Coding System (FACS), essential for understanding a wide range of realistic facial movements and enhancing digital characters' emotional depth and reliability. A novel innovation of the thesis is a tool that offers animators control over the range of movement of facial expressions. This tool's parametric functionality enables detailed, incremental adjustments, capturing human emotions' subtleties in digital form. The study demonstrates a systematic approach to developing a Parametric Framework and Tool for posing photorealistic 3D character facial expressions. It employs a Design Science Research Methodology, focusing on practical solution-oriented research. Key stages include problem identification, defining objectives, design and development of the tool, demonstration, evaluation, and communication. The process is iterative, integrating knowledge from facial expression theory, animation, computer graphics, and user interface design. The methodology emphasises user testing with animators and learners, incorporating feedback to refine the tool, ensuring practicality and utility for its intended users. The thesis emphasises successfully developing a Parametric Framework and Tool and highlights advancements in bridging the gap between theoretical knowledge and practical application in facial animation. The thesis concludes by underscoring the potential impact of this work on the animation industry, improving the emotional depth and realism of digital characters. Future directions for research and development are suggested, focusing on further enhancing the tool's capabilities and exploring new applications in various digital media contexts.

History

Qualification name

  • PhD

Supervisor

Wall, Darren ; Bowen, Sarah

Awarding Institution

Leeds Beckett University

Completion Date

2024-10-01

Qualification level

  • Doctoral

Language

  • eng

Publisher

Leeds Beckett University

Usage metrics

    LBU Theses and Dissertations

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC