Burley’s article (2012) on a material model developed at Disney for feature films is an excellent read. It includes extensive discussion of features of real-world reflection functions that can be observed in Matusik et al.’s (2003b) measurements of one hundred BRDFs and analyzes the ways that existing BRDF models do and do not fit these features well. These insights are then used to develop an “artist-friendly” material model that can express a wide range of surface appearances. The model describes reflection with a single color and ten scalar parameters, all of which are in the range and have fairly predictable effects on the appearance of the resulting material.
Blinn (1978) invented the bump-mapping technique. Kajiya (1985) generalized the idea of bump mapping the normal to frame mapping, which also perturbs the surface’s primary tangent vector and is useful for controlling the appearance of anisotropic reflection models. Mikkelsen’s thesis (2008) carefully investigates a number of the assumptions underlying bump mapping, proposes generalizations, and addresses a number of subtleties with respect to its application to real-time rendering.
Snyder and Barr (1987) noted the light leak problem from per-vertex shading normals and proposed a number of work-arounds. The method we have used in this chapter is from Section 5.3 of Veach’s thesis (1997); it is a more robust solution than those of Snyder and Barr.
Shading normals introduce a number of subtle problems for physically based light transport algorithms that we have not addressed in this chapter. For example, they can easily lead to surfaces that reflect more energy than was incident upon them, which can wreak havoc with light transport algorithms that are designed under the assumption of energy conservation. Veach (1996) investigated this issue in depth and developed a number of solutions. Section 16.1 of this book will return to this issue.
One visual shortcoming of bump mapping is that it doesn’t naturally account for self-shadowing, where bumps cast shadows on the surface and prevent light from reaching nearby points. These shadows can have a significant impact on the appearance of rough surfaces. Max (1988) developed the horizon mapping technique, which performs a preprocess on bump maps stored in image maps to compute a term to account for this effect. This approach isn’t directly applicable to procedural textures, however. Dana et al. (1999) measured spatially varying reflection properties from real-world surfaces, including these self-shadowing effects; they convincingly demonstrate this effect’s importance for accurate image synthesis.
Another difficult issue related to bump mapping is that antialiasing bump maps that have higher frequency detail than can be represented in the image is quite difficult. In particular, it is not enough to remove high-frequency detail from the bump map function, but in general the BSDF needs to be modified to account for this detail. Fournier (1992) applied normal distribution functions to this problem, where the surface normal was generalized to represent a distribution of normal directions. Becker and Max (1993) developed algorithms for blending between bump maps and BRDFs that represented higher-frequency details. Schilling (1997, 2001) investigated this issue particularly for application to graphics hardware. More recently, effective approaches to filtering bump maps were developed by Han et al. (2007) and Olano and Baker (2010). Recent work by Dupuy et al. (2013) and Hery et al. (2014) addressed this issue by developing techniques that convert displacements into anisotropic distributions of Beckmann microfacets.
An alternative to bump mapping is displacement mapping, where the bump function is used to actually modify the surface geometry, rather than just perturbing the normal (Cook 1984; Cook et al. 1987). Advantages of displacement mapping include geometric detail on object silhouettes and the possibility of accounting for self-shadowing. Patterson and collaborators described an innovative algorithm for displacement mapping with ray tracing where the geometry is unperturbed but the ray’s direction is modified such that the intersections that are found are the same as would be found with the displaced geometry (Patterson et al. 1991; Logie and Patterson 1994). Heidrich and Seidel (1998) developed a technique for computing direct intersections with procedurally defined displacement functions.
As computers have become faster, another viable approach for displacement mapping has been to use an implicit function to define the displaced surface and to then take steps along rays until they find a zero crossing with the implicit function. At this point, an intersection has been found. This approach was first introduced by Hart (1996); see Donnelly (2005) for information about using this approach for displacement mapping on the GPU. This approach was recently popularized by Quilez on the shadertoy Web site (Quilez 2015).
With the advent of increased memory on computers and caching algorithms, the option of finely tessellating geometry and displacing its vertices for ray tracing has become feasible. Pharr and Hanrahan (1996) described an approach to this problem based on geometry caching, and Wang et al. (2000) described an adaptive tessellation algorithm that reduces memory requirements. Smits, Shirley, and Stark (2000) lazily tessellate individual triangles, saving a substantial amount of memory.
Measuring fine-scale surface geometry of real surfaces to acquire bump or displacement maps can be challenging. Johnson et al. (2011) developed a novel hand-held system that can measure detail down to a few microns, which more than suffices for these uses.
- Amanatides, J. 1992. Algorithms for the detection and elimination of specular aliasing. In Proceedings of Graphics Interface 1992, 86–93.
- Becker, B. G., and N. L. Max. 1993. Smooth transitions between bump rendering algorithms. In Proceedings of SIGGRAPH ’93, Computer Graphics Proceedings, Annual Conference Series, 183–90.
- Blinn, J. F. 1978. Simulation of wrinkled surfaces. In Computer Graphics (SIGGRAPH ’78 Proceedings), 12, 286–92.
- Burley, B. Physically-based shading at Disney. Physically Based Shading in Film and Game Production, SIGGRAPH 2012 Course Notes.
- Cook, R. L. 1984. Shade trees. Computer Graphics (SIGGRAPH ’84 Proceedings), 18, 223–31.
- Cook, R. L., L. Carpenter, and E. Catmull. 1987. The Reyes image rendering architecture. Computer Graphics (Proceedings of SIGGRAPH ’87), 95–102.
- Dana, K. J., B. van Ginneken, S. K. Nayar, and J. J. Koenderink. 1999. Reflectance and texture of real-world surfaces. ACM Transactions on Graphics 18 (1), 1–34.
- Donnelly, W. Per-pixel displacement mapping with distance functions. In M. Pharr (Ed.), GPU Gems 2. Reading, Massachusetts: Addison-Wesley.
- Dupuy, J., E. Heitz, J.-C. Iehl, P. Poulin, F. Neyret, and V. Ostromoukhov. Linear efficient antialiased displacement and reflectance mapping. ACM Transactions on Graphics 32 (6).
- Fisher, M., K. Fatahalian, S. Boulos, K. Akeley, W. R. Mark, and P. Hanrahan. DiagSplit: parallel, crack-free, adaptive tessellation for micropolygon rendering. ACM Transactions on Graphics (Proceedings of ACM SIGGRAPH Asia 2009) 28 (5), 150:1–150:10.
- Fournier, A. 1992. Normal distribution functions and multiple surfaces. In Graphics Interface ’92 Workshop on Local Illumination, 45–52.
- Han, C., B. Sun, R. Ramamoorthi, and E. Grinspun. 2007. Frequency domain normal map filtering. ACM Transactions on Graphics (Proceedings of SIGGRAPH 2007) 26 (3), 28:1–28:11.
- Hart, J. C. 1996. Sphere tracing: a geometric method for the antialiased ray tracing of implicit surfaces. The Visual Computer 12 (9), 527–45.
- Heidrich, W., and H.-P. Seidel. 1998. Ray-tracing procedural displacement shaders. In Proceedings of Graphics Interface 1998, 8–16.
- Heitz, E., and F. Neyret. Representing appearance and pre-filtering subpixel data in sparse voxel octrees. In Proceedings of High Performance Graphics 2012, 125–34.
- Heitz, E., J. Dupuy, C. Crassin, and C. Dachsbacher. The SGGX microflake distribution. ACM Transactions on Graphics (Proceedings of SIGGRAPH 2015) 34 (4), 48:1–48:11.
- Hery, C., M. Kass, and J. Ling. Geometry into shading. Pixar Technical Memo 14-04.
- Johnson, M. K., F. Cole, A. Raj, and E. H. Adelson. Microgeometry capture using an elastomeric sensor. ACM Transactions on Graphics (Proceedings of SIGGRAPH 2011) 30 (4), 46:1–46:8.
- Kajiya, J. T. 1985. Anisotropic reflection models. Computer Graphics (Proceedings of SIGGRAPH ’85), 19, 15–21.
- Logie, J. R., and J. W. Patterson. 1994. Inverse displacement mapping in the general case. Computer Graphics Forum 14 (5), 261–73.
- Matusik, W., H. Pfister, M. Brand, and L. McMillan. 2003b. A data-driven reflectance model. ACM Transactions on Graphics (Proceedings of SIGGRAPH 2003) 22 (3), 759–69.
- Max, N. L. 1988. Horizon mapping: shadows for bump-mapped surfaces. The Visual Computer 4 (2), 109–17.
- Mikkelsen, M. 2008. Simulation of wrinkled surfaces revisited. M.S. thesis, University of Copenhagen.
- Neyret, F. 1996. Synthesizing verdant landscapes using volumetric textures. In Eurographics Rendering Workshop 1996, 215–24.
- Neyret, F. 1998. Modeling, animating, and rendering complex scenes using volumetric textures. IEEE Transactions on Visualization and Computer Graphics 4 (1), 55–70.
- Olano, M., and D. Baker. LEAN mapping. In Proceedings of the 2010 ACM SIGGRAPH symposium on Interactive 3D Graphics and Games, 181–88.
- Patney, A., M. S. Ebeida, and J. D. Owens. Parallel view-dependent tessellation of Catmull–Clark subdivision surfaces. In Proceedings of High Performance Graphics 2009, 99–108.
- Patterson, J. W., S. G. Hoggar, and J. R. Logie. 1991. Inverse displacement mapping. Computer Graphics Forum 10 (2), 129–39.
- Pharr, M., and P. Hanrahan. 1996. Geometry caching for ray-tracing displacement maps. In Eurographics Rendering Workshop 1996, 31–40.
- Quilez, I. Distance estimation. http://iquilezles.org/www/articles/distance/distance.htm.
- Schilling, A. 1997. Toward real-time photorealistic rendering: challenges and solutions. In 1997 SIGGRAPH/Eurographics Workshop on Graphics Hardware, 7–16.
- Schilling, A. 2001. Antialiasing of environment maps. Computer Graphics Forum 20 (1), 5–11.
- Smits, B., P. S. Shirley, and M. M. Stark. 2000. Direct ray tracing of displacement mapped triangles. In Rendering Techniques 2000: 11th Eurographics Workshop on Rendering, 307–18.
- Snyder, J. M., and A. H. Barr. 1987. Ray tracing complex models containing surface tessellations. Computer Graphics (SIGGRAPH ’87 Proceedings), Volume 21, 119–28.
- Veach, E. 1996. Non-symmetric scattering in light transport algorithms. In X. Pueyo and P. Schröder (Eds.), Eurographics Rendering Workshop 1996. Wien: Springer.
- Veach, E. 1997. Robust Monte Carlo methods for light transport simulation. Ph.D. thesis, Stanford University.
- Wang, X. C., J. Maillot, E. L. Fiume, V. Ng-Thow-Hing, A. Woo, and S. Bakshi. 2000. Feature-based displacement mapping. In Rendering Techniques 2000: 11th Eurographics Workshop on Rendering, 257–68.