Use this URL to cite or link to this record in EThOS:
Title: GPU-based volume deformation
Author: Walton, Simon
Awarding Body: Swansea University
Current Institution: Swansea University
Date of Award: 2007
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Surface-based representations of objects, and consequently their rendering algorithms, currently dominate the field of computer graphics. It could be argued that this is not just due to the efficiency of representation (representing merely surfaces, and not internal information), but is mostly due to the fact that surface-based graphics as a sub-field has seen many years of prioritised research and development. Volume graphics as a sub-field of computer graphics has however seen a rapid rise in research concentration in recent years. Its popularity can be attributed mainly to its ever-important role in medical applications such as surgery simulations and medical illustration; however, its rapid growth in the past five years or so is unquestionably due to the real-time volume rendering techniques implemented on the Graphics Processing Units of commodity graphics hardware. The deformation of graphical objects is an important part of animation; particularly in CGI-based movies where characters must bend and stretch comically according to their actions. Deformation also plays an important role in surgical simulations, where real-time physically-based solutions are required to give the surgeon or student a realistic simulation of a surgical operation. The deformation of volumetric data (as in volume graphics) is a challenge due to the sheer amount of data that must be transformed, and the lack of topographical/semantic information that is embedded with freshly-aquired data. Such semantics must usually be inferred by the user using manual processes such as segmentation. The work presented in this thesis provides a robust set of methods and techniques for the real-time manipulation of volumetric data, utilising high-performance graphics hardware to ensure that the field of volume graphics can continue to be a highly-attractive alternative to surface-based graphics. The main contributions of this work are: • A comprehensive review of volume graphics and volume deformation; • An introduction to important GPU-acclererated volume graphics methods; • A framework for the non-reconstructive deformation of volume data; • A GPU-accelerated forward-projection system for interactive volume deformation; • A real-time backwardbackward-mapping raycasting Tenderer for interactive, character-based volume deformation.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available