Modular Radiance Transfer

晃晃

Modular Radiance Transfer

Bradford J. Loos1;2  Lakulish Antani1;3 Kenny Mitchell 1 Derek Nowrouzezahrai 4 Wojciech Jarosz4 Peter-Pike Sloan1

1Disney Interactive Studios 2University of Utah 3UNC Chapel Hill 4Disney Research Zurich









Abstract

Many rendering algorithms willingly sacrifice accuracy, favoring

plausible shading with high-performance. Modular Radiance Trans-

fer (MRT) models coarse-scale, distant indirect lighting effects in

scene geometry that scales from high-end GPUs to low-end mobile

platforms. MRT eliminates scene-dependent precomputation by

storing compact transport on simple shapes, akin to bounce cards

used in film production. These shapes’ modular transport can be in-

stanced, warped and connected on-the-fly to yield approximate light

transport in large scenes. We introduce a prior on incident lighting

distributions and perform all computations in low-dimensional sub-

spaces. An implicit lighting environment induced from the low-rank

approximations is in turn used to model secondary effects, such as

volumetric transport variation, higher-order irradiance, and transport

through lightfields. MRT is a new approach to precomputed lighting

that uses a novel low-dimensional subspace simulation of light trans-

port to uniquely balance the need for high-performance and portable

solutions, low memory usage, and fast authoring iteration.

CR Categories: I.3.7 [Computer Graphics]: Three-Dimensional

Graphics and Realism—Color, shading, shadowing, and texture;

Keywords: Global Illumination, GPU, Interactive

1 Introduction

Indirect illumination increases the realism of computer generated

images. The ambient term is a simple inexpensive approximation

that does not respond to dynamic lighting. Accurate real-time tech-

niques [Keller 1997] have difficulty scaling to complex scenes and

often have significant performance requirements, particularly on

modern console and mobile platforms. Techniques that approximate

different elements of indirect lighting have been extremely success-

ful in interactive graphics applications. Precomputation techniques

used in video games [Chen 2008; Larsson and Halen 2009] tend to

assume static scenes and lighting, but suffer from long authoring

iteration times and memory requirements. Ambient Occlusion (AO)

[Zhukov et al. 1998] captures only salient shading effects. Variants

of Precomputed Radiance Transfer (PRT) [Sloan et al. 2002] gener-

ate soft lighting results. These techniques are favorable compared to

more accurate techniques due to their lower storage and computation

costs and the pleasing nature of their approximation. Modular Radi-

ance Transfer (MRT) targets coarse-scale, distant indirect lighting

in scene geometry, responds plausibly and smoothly to dynamic

lighting, has extremely high-performance, and allows fast author

iteration.

Our shapes are motivated by bounce cards used in live-action films.

These planes only approximate indirect light from geometry in the

real-world but offer a high level of control to produce the desired

lighting. In digital film production, non-shadow casting lights are

commonly used to allow artists to quickly iterate and achieve a

desired look. The ease-of-use and controllability of these approxi-

mations outweighs their physically incorrect nature.

Our approach is very efficient, uses very little data and a quick, one-

time, scene-independent precomputation step. It also allows real-

time computation of approximate indirect light, and is designed with

rapid iteration of light design in mind. We precompute light transport

operators (LTOs) for a handful of simple canonical “shapes”, then

interactively warp and combine these shapes, along with their LTOs,

to more complex geometry. These shape proxies are used to model

direct-to-indirect transport which is then applied as a light map to the

actual scene geometry. The flow of indirect light between proxies

is modeled with lightfields, and all computations are performed on

very low-dimensional subspaces. MRT results in plausible, dynamic

global-illumination effects, rendered at high frame rates with low

memory overhead. We design special LTOs for secondary transport

effects such as light volumes for dynamic characters and higher-

order irradiance for normal mapping. We illustrate our solutions

ability to scale from high-end to mobile platforms and, like PRT, to

provide smooth results which respond to light change.









全文请下载附件：

菜刀吻电线

有意思！学习了！

晃晃

水……生命之源……灌……

奇

凡系斑竹滴话要听；凡系朋友滴帖要顶

tc

加精、加亮滴铁子，尤其要多丁页丁页

C.R.CAN

我看看就走，你们聊！

奇

呵呵,真得不错哦!!

晃晃

凡系斑竹滴话要听；凡系朋友滴帖要顶

铁锹 3D打印店铺

[速递]3D打印概念涨3.02% 光韵达涨停



半虚拟3D恐怖生存《Montas》




3D打印风行背后的三种兴奋和三种思考




迪斯尼3D打印公主和交互蛋糕




BI虚拟现实_商业分析更助决策




增强现实与潮流文化




博物馆网上争开“虚拟现实馆”




虚拟现实3D云技术3D数据压缩比150:1




盗梦空间：头盔试验者难辨现实与虚幻




动漫“虚拟制造”动起来




闲谈“虚拟现实”_时下新宠技术

C.R.CAN

都闪开，介个帖子，偶来顶