Photo‐realistic Rendering of Metallic Car Paint from Image‐Based Measurements

Abstract

Figure 1: Based on analytical reflectance modeling and image-based rendering our technique accurately reproduces complex effects of modern car paint like specular reflection, spatially varying glitter with depth impression and color shifts. AbstractState-of-the-art car paint shows not only interesting and subtle angular dependency but also significant spatial variation. Especially in sunlight these variations remain visible even for distances up to a few meters and give the coating a strong impression of depth which cannot be reproduced by a single BRDF model and the kind of procedural noise textures typically used. Instead of explicitly modeling the responsible effect particles we propose to use image-based reflectance measurements of real paint samples and represent their spatial varying part by Bidirectional Texture Functions (BTF). We use classical BRDF models like Cook-Torrance to represent the reflection behavior of the base paint and the highly specular finish and demonstrate how the parameters of these models can be derived from the BTF measurements. For rendering, the image-based spatially varying part is compressed and efficiently synthesized. This paper introduces the first hybrid analytical and image-based representation for car paint and enables the photo-realistic rendering of all significant effects of highly complex coatings.

Discrete stochastic microfacet models

“…This work has identified many salient perceptual effects, including the texture characteristics of diffuse coarseness (contrast of the noise seen under diffuse illumination) and glints (bright points seen under unidirectional illumination). Cars are also an important rendering application, and several models for realistic rendering of metallic car paints have been proposed that handle glitter using procedural normal maps [Günther et al 2005], measured textures [Rump et al 2008], or noise patterns [Ershov et al 1999;Ershov et al 2001].…”

A Reflectance Model for Metallic Paints Using a Two-Layer Structure Surface with Microfacet Distributions

“…For model fitting methods, several researchers including Günter et al [5] and Rump et al [7] have fitted the measured data to a Cook-Torrance model with 2 or 3 lobes to represent the homogeneous BRDF part of car paints, but their research basically follows the study of Ngan et al [6]. Ngan analyzed various analytical models using Matusik's 100 isotropic BRDF data sets which include metallic paints.…”

“…To evaluate the performance of our model, we have selected a Cook-Torrance model with 2 specular lobes and the Ershov's model which have been used for representing metallic paints based on measured data in previous studies [5], [7], [20]. We test the two-layer model described in Sect.…”

“…Recently, modeling methods ensuring the accuracy requirement of appearance materials by measuring the reflectance characteristics have been developed. These measurement-based methods can be classified into two types: factorization methods [2]- [4] and measure-and-fit methods [5]- [7]. A factorization method decomposes huge and highdimensional measured data into sub-dimensional basis functions using a decomposition technique such as PCA (principal component analysis), SVD (singular value decomposition) and NMF (non-negative matrix factorization).…”

“…Since analytical models can express only the phenomena which are considered in the models, it is essential to select a model that properly represents the reflectance characteristics of the target material. As an analytical model for metallic paints, either specialized models for paint coatings like the Ershov model [8] or general models for opaque surfaces like the Cook-Torrance [9] with multiple lobes can be used [7].…”