Compensation compliant appearance editing of physical objects with arbitrary shape and color
Abstract
Altering the visual content on an arbitrary physical surface or set of objects, or appearance editing, is an important function for many computer graphics applications. A desired appearance can be accomplished by compensating the physical object's surfaces with controlled light from a set of one or more digital projectors. When the desired appearance is achievable, it is said to be compensation compliant. However, the range of compensation compliant colors in an appearance edit is limited by the surface's orientation to the projectors and the surface's reflectance properties. Incorrectly compensated colors result in color shifts which may produce misleading visualizations. This dissertation formulates compensation compliancy and proposes new approaches to mitigate limitations to achieving compensation compliant appearance edits on diffuse surfaces. First, luminance compliancy is discussed in the framework of a virtual restoration system which visually restores the appearance of deteriorated artifacts to their original appearance. Then, in addition to luminance, a color's chrominance is considered in a formulation of compensation compliancy in RGB color space. An automatic projector planning algorithm is discussed which computes optimal locations to place projectors in order to achieve a desired target appearance. The modeling of compliancy is further improved with a formulation in CIELAB color space, and an approach to automatically achieving compliancy by minimally altering a target appearance is discussed. Finally, a new paradigm of appearance editing with present environment light is discussed. This paradigm generalizes the formulation of compensation compliancy in CIELAB color space to its most flexible and accurate form. Altogether, these approaches to improving compliancy can be combined into a comprehensive framework to automatically achieve compensation compliant appearance edits.
Degree
Ph.D.
Advisors
Aliaga, Purdue University.
Subject Area
Computer science
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.