Abstract: This paper presents a hybrid algorithm for rendering approximate motion and defocus blur with precise stochastic visibility evaluation. It demonstrates---for the first time, with a full stochastic technique---real-time performance on conventional GPU architectures for complex scenes at 1920x1080 HD resolution. The algorithm operates on dynamic triangle meshes for which per-vertex velocity or corresponding vertices from the previous frame are available. It leverages multisample antialiasing (MSAA) and a tight space-time-aperture convex hull to efficiently evaluate visibility independently of shading. For triangles that cross z=0, it fall backs to a 2D bounding box that we hypothesize but do not prove is conservative. The algorithm further reduces sample variance within primitives by integrating textures according to ray differentials in time and aperture.

Abstract: Stochastic transparency provides a unified approach to order-independent transparency, anti-aliasing, and deep shadow maps. It augments screen-door transparency using a random sub-pixel stipple pattern, where each fragment of transparent geometry covers a random subset of pixel samples of size proportional to alpha. This results in correct alpha-blended colors on average, in a single render pass with fixed memory size and no sorting, but introduces noise. We reduce this noise by an alpha correction pass, and by an accumulation pass that uses a stochastic shadow map from the camera. At the pixel level, the algorithm does not branch and contains no read-modify-write loops other than traditional z-buffer blend operations. This makes it an excellent match for modern massively parallel GPU hardware. Stochastic transparency is very simple to implement and supports all types of transparent geometry, able without coding for special cases to mix hair, smoke, foliage, windows, and transparent cloth in a single scene.

Abstract: Existing offline techniques for modeling subsurface scattering effects in multi-layered translucent materials such as human skin achieve remarkable realism, but require seconds or minutes to generate an image. We demonstrate rendering of multi-layer skin that achieves similar visual quality but runs orders of magnitude faster. We show that sums of Gaussians provide an accurate approximation of translucent layer diffusion profiles, and use this observation to build a novel skin rendering algorithm based on texture space diffusion and translucent shadow maps. Our technique requires a parameterized model but does not otherwise rely on any precomputed information, and thus extends trivially to animated or deforming models. We achieve about 30 frames per second for realistic real-time rendering of deformable human skin under dynamic lighting.

Abstract: High-quality off-line rendering requires many features not natively supported by current commodity graphics hardware: wide smooth filters, high sampling rates, order-independent transparency, spectral opacity, motion blur, depth of field. We present a GPU-based hidden-surface algorithm that implements all these features. The algorithm is Reyes-like but uses regular sampling and multiple passes. Transparency is implemented by depth peeling, made more efficient by opacity thresholding and a new method called \emph{z batches}. We discuss performance and some design trade-offs. At high spatial sampling rates, our implementation is substantially faster than a CPU-only renderer for typical scenes.

Abstract: Finely detailed 3D geometry can show significant aliasing artifacts if rendered using native hardware multisampling, because multisampling is currently limited to one-pixel box filtering and low sampling rates. This chapter describes a tiled supersampling technique for rendering images of arbitrary resolution with arbitrarily wide user-defined filters and high sampling rates. The code presented here is used in the Gelato film renderer to produce images of uncompromising quality using the GPU.