Physically-based spectral rendering is becoming increasingly popular in both commercial and academic areas due to its ability to accurately simulate natural phenomena. However, the production of materials defined by their spectral properties is a tedious and expensive process, which makes the utilization of RGB-based assets in spectral renderers a desired feature. To convert RGB values to their spectral representations, a process called spectral uplifting is employed. As the RGB color space is a finite subset of the visible gamut, there exist multiple conversion techniques producing distinct results, which may cause color inconsistencies under various lighting conditions. This thesis proposes a method for constraining the spectral uplifting process. To be specific, pre-defined mappings of RGB values to their spectral representations are preserved and the rest of the RGB gamut is plausibly uplifted. In order to assess its correctness, this technique is then implemented and evaluated in a spectral renderer. The renders uplifted via our method show minimal discrepancies when compared to the original textures.

No / Visual psychophysicists, who study object, color, and light perception, have a demand for software that produces complex but, at the same time, physically accurate stimuli for their experiments. The number of computer graphic packages that simulate the physical interaction of light and surfaces is limited, and mostly they require the purchase of a license. RADIANCE (Ward, 1994), however, is freely available and popular in the visual perception community, making it a prime candidate. We have shown previously that RADIANCE¿s simulation accuracy is greatly improved when color is coded by spectra, rather than by the originally envisaged RGB triplets (Ruppertsberg & Bloj, 2006). Here, we present a method for spectral rendering with RADIANCE to generate hyperspectral images that can be converted to XYZ images (CIE 1931 system) and then to machine-dependent RGB images. Generating XYZ stimuli has the added advantage of making stimulus images independent of display devices and, thereby, facilitating the process of reproducing results across different labs. Materials associated with this article may be downloaded from www.psychonomic.org.

Various rendering techniques often use different approaches to the same aspects of the image synthesis process, mainly due to their complexity and constant development. Excluding global illumination algorithms, appearance descriptions are key distinguishing factors between the rendering systems. These descriptions might include BRDF models, support for spectral color representation, and even integration of advanced phenomena, such as fluores- cence. Unfortunately, as there are no standardized implementations of these features, their computations might not be completely accurate, which may result in their incorrect representation. This thesis describes an evaluation suite that methodically tests rendering algorithms based on their appearance reproduction capabilities. The core of the suite is a set of scenes that test five specific appearance phenomena - polarization, GGX reflectance, fluorescence, iridescence and the overall spectral accuracy. Each test case scenario contains as few scenes as possible while maximizing the number of covered aspects of the tested feature. For the user's convenience, we wrap the scenes inside an automatic workflow that runs the specified test case scenarios and displays the results. As a correctness metric, we provide manually verified reference images that are considered to...

Spectral rendering, or the synthesis of images by taking into account the wavelengths of light, allows effects otherwise impossible with other methods. One of these effects is dispersion, the phenomenon that creates a rainbow when white light shines through a prism. Spectral rendering has previously remained in the realm of off-line rendering (with a few exceptions) due to the extensive computation required to keep track of individual light wavelengths. Caustics, the focusing and de-focusing of light through a refractive medium, can be interpreted as a special case of dispersion where all the wavelengths travel together. This thesis extends Adaptive Caustic Mapping, a previously proposed caustics mapping algorithm, to handle spectral dispersion. Because ACM can display caustics in real-time, it is quite amenable to be extended to handle the more general case of dispersion. A method is presented that runs in screen-space and is fast enough to display plausible dispersion phenomena in real-time at interactive frame rates.

Within the last decade, the offline rendering branch of computer graphics has moved towards the concept of physically-based rendering by using the path tracing algorithm. One such physically-based effect is fluorescence, where light is absorbed at one wavelength and re-emitted at another. However, to properly capture this effect, one has to utilize spectral path tracing, as opposed to colour- based path tracing. Spectral path tracing by itself suffers from colour noise, which can be reduced by utilizing the so-called Hero Wavelength Spectral Sampling (HWSS). The inclusion of wavelength shifting induced by fluorescence requires modifications to the base path tracing algorithm that HWSS wasn't designed for. This thesis provides the overview of path tracing, the current state-of-art for in- clusion of fluorescence in a rendering system, along with relevant technical details, the overview of HWSS itself as well as mathematical formulation that enables the combination of fluorescence and HWSS. Additionally, this thesis also proposes a new approach to rendering fluorescent participating media that properly handles previously overlooked failure cases. 1

Dans le domaine de la synthèse d'image, la simulation de l'apparence visuelle des matériaux nécessite, la résolution rigoureuse de l'équation du transport de la lumière. Cela implique d'incorporer dans les modèles tous les éléments pouvant avoir une influence sur la luminance spectrale énergétique reçue par l'œil humain. La caractérisation des propriétés de réflectance des matériaux, encore sujette à de nombreuses recherches, est très évoluée. Cependant, l'utilisation de cartes d'environnement, pour simuler leurs comportements visuels restent essentiellement trichromatiques. Caractériser la lumière naturelle avec précision, est une interrogation ancienne et il n'existe pas aujourd'hui de cartes d'environnement comportant à la fois les informations de luminance spectrale énergétique et de polarisations correspondant à des ciels réels. Il nous est donc apparu nécessaire, de proposer à la communauté de l'informatique graphique des environnements lumineux complets exploitables dans un moteur de rendu adapté en conséquence.Dans ce travail, nous exploitons des résultats issus d'autres domaines scientifiques tels que la météorologie, la climatologie..., pour proposer un modèle de ciel clair, c'est-à-dire sans nuage.Toutes les situations réelles ne pouvant pas être abordées par cette méthode, nous développons et caractérisons un dispositif de capture d'environnement lumineux incorporant à la fois, la gamme dynamique de l'éclairage, la répartition spectrale et les états de polarisation.Nous proposons, dans le but de standardiser les échanges, un format de données utilisable dans un moteur de rendu spectral, exploitant le formalisme de "Stokes - Mueller". / In the field of computer graphics, the simulation of the visual appearance of materials requires, a rigorous solving of the light transport equation. This implies to incorporate into models all elements that can influence the spectral received by human eyes. The characterization of the reflectance properties of materials, still subject to many researches is very advanced. However, the uses of environment maps, to simulate their visual behaviors remain essentially trichromaticity. Characterize the natural light with precision, is an old question. Today, there are no environment maps, including both spectral radiance and polarization informations, corresponding to a real sky. It was therefore necessary for us to design and propose to the computer graphics community a full of bright environments exploitable in a rendering engine adapted accordingly. In this work, we use the results of other scientific fields as meteorology, climatology..., to propose a new model of clear sky. As all actual situations are not addressed by this method, we develop and characterize an environment capturing device both incorporating the light dynamic range, the spectral distribution and the polarization states.

In computer graphics, the complex phenomenon of color appearance, involving the interaction of light, matter and the human visual system, is modeled by the multiplication of RGB triplets assigned to lights and materials. This efficient heuristic produces plausible images because the triplets assigned to materials usually function as color specifications. To predict color, spectral rendering is required, but the O(n) cost of computing reflections with n-dimensional point-sampled spectra is prohibitive for real-time rendering. Typical spectra are well approximated by m-dimensional linear models, where m << n, but computing reflections with this representation requires O(m^2) matrix-vector multiplication. A method by Drew and Finlayson [JOSA A 20, 7 (2003), 1181-1193], reduces this cost to O(m) by “sharpening” an n x m orthonormal basis with a linear transformation, so that the new basis vectors are approximately disjoint. If successful, this transformation allows approximated reflections to be computed as the products of coefficients of lights and materials. Finding the m x m change of basis matrix requires solving m eigenvector problems, each needing a choice of wavelengths in which to sharpen the corresponding basis vector. These choices, however, are themselves an optimization problem left unaddressed by the method's authors. Instead, we pose a single problem, expressing the total approximation error incurred across all wavelengths as the sum of dm^2 squares for some number d, where, depending on the inherent dimensionality of the rendered reflectance spectra, m <= d << n, a number that is independent of the number of approximated reflections. This problem may be solved in real time, or nearly, using standard nonlinear optimization algorithms. Results using a variety of reflectance spectra and three standard illuminants yield errors at or close to the best lower bound attained by projection onto the leading m characteristic vectors of the approximated reflections. Measured as CIEDE2000 color differences, a heuristic proxy for image difference, these errors can be made small enough to be likely imperceptible using values of 4 <= m <= 9. An examination of this problem reveals a hierarchy of simpler, more quickly solved subproblems whose solutions yield, in the typical case, increasingly inaccurate approximations. Analysis of this hierarchy explains why, in general, the lowest approximation error is not attained by simple spectral sharpening, the smallest of these subproblems, unless the spectral power distributions of all light sources in a scene are sufficiently close to constant functions. Using the methods described in this dissertation, spectra can be rendered in real time as the products of m-dimensional vectors of sharp basis coefficients at a cost that is, in a typical application, a negligible fraction above the cost of RGB rendering. / Graduate