Results 1  10
of
35
Instant Radiosity
, 1997
"... We present a fundamental procedure for instant rendering from the radiance equation. Operating directly on the textured scene description, the very efficient and simple algorithm produces photorealistic images without any finite element kernel or solution discretization of the underlying integral eq ..."
Abstract

Cited by 180 (3 self)
 Add to MetaCart
We present a fundamental procedure for instant rendering from the radiance equation. Operating directly on the textured scene description, the very efficient and simple algorithm produces photorealistic images without any finite element kernel or solution discretization of the underlying integral equation. Rendering rates of a few seconds are obtained by exploiting graphics hardware, the deterministic technique of the quasirandom walk for the solution of the global illumination problem, and the new method of jittered low discrepancy sampling.
Interactive rendering of large unstructured grids using dynamic levelofdetail
 In IEEE Visualization ’05
, 2005
"... We describe a new dynamic levelofdetail (LOD) technique that allows realtime rendering of large tetrahedral meshes. Unlike approaches that require hierarchies of tetrahedra, our approach uses a subset of the faces that compose the mesh. No connectivity is used for these faces so our technique eli ..."
Abstract

Cited by 33 (11 self)
 Add to MetaCart
We describe a new dynamic levelofdetail (LOD) technique that allows realtime rendering of large tetrahedral meshes. Unlike approaches that require hierarchies of tetrahedra, our approach uses a subset of the faces that compose the mesh. No connectivity is used for these faces so our technique eliminates the need for topological information and hierarchical data structures. By operating on a simple set of triangular faces, our algorithm allows a robust and straightforward graphics hardware (GPU) implementation. Because the subset of faces processed can be constrained to arbitrary size, interactive rendering is possible for a wide range of data sets and hardware configurations.
Fast and Accurate Hierarchical Radiosity Using Global Visibility
, 1999
"... this paper we present a new algorithm which addresses the three shortcomings mentioned above. For all three problems, refinement, meshing and visibility previous approaches lack information on accurate global visibility relationships in the scene. This information is provided by the Visibility Skele ..."
Abstract

Cited by 29 (2 self)
 Add to MetaCart
this paper we present a new algorithm which addresses the three shortcomings mentioned above. For all three problems, refinement, meshing and visibility previous approaches lack information on accurate global visibility relationships in the scene. This information is provided by the Visibility Skeleton [Durand et al. 1997]. To achieve our goal, we first extend the Skeleton to provide visibility information at vertices resulting from subdivision of the original input surfaces. The extended Skeleton allows the fast computation of exact pointtopolygon formfactors for any pointpolygon pair in the scene. In addition, all visibility information (blockers and all discontinuity surfaces) is available for any polygonpolygon pair.
Efficient Multidimensional Sampling
, 2002
"... Image synthesis often requires the Monte Carlo estimation of integrals. Based on a generalized concept of stratification we present an efficient sampling scheme that consistently outperforms previous techniques. This is achieved by assembling sampling patterns that are stratified in the sense of jit ..."
Abstract

Cited by 25 (1 self)
 Add to MetaCart
Image synthesis often requires the Monte Carlo estimation of integrals. Based on a generalized concept of stratification we present an efficient sampling scheme that consistently outperforms previous techniques. This is achieved by assembling sampling patterns that are stratified in the sense of jittered sampling and Nrooks sampling at the same time. The faster convergence and improved antialiasing are demonstrated by numerical experiments.
Fourier Depth of Field
"... Optical systems used in photography and cinema produce depth of field effects, that is, variations of focus with depth. These effects are simulated in image synthesis by integrating incoming radiance at each pixel over the lense aperture. Unfortunately, aperture integration is extremely costly for d ..."
Abstract

Cited by 20 (7 self)
 Add to MetaCart
Optical systems used in photography and cinema produce depth of field effects, that is, variations of focus with depth. These effects are simulated in image synthesis by integrating incoming radiance at each pixel over the lense aperture. Unfortunately, aperture integration is extremely costly for defocused areas where the incoming radiance has high variance, since many samples are then required for a noisefree Monte Carlo integration. On the other hand, using many aperture samples is wasteful in focused areas where the integrand varies little. Similarly, image sampling in defocused areas should be adapted to the very smooth appearance variations due to blurring. This paper introduces an analysis of focusing and depth of field in the frequency domain, allowing a practical characterization of a light field’s frequency content both for image and aperture sampling. Based on this analysis we propose an adaptive depth of field rendering algorithm which optimizes sampling in two important ways. First, image sampling is based on conservative bandwidth prediction and a splatting reconstruction technique ensures correct image reconstruction. Second, at each pixel the variance in the radiance over the aperture is estimated, and used to govern sampling. This technique is easily integrated in any samplingbased renderer, and vastly improves performance.
Hierarchical Monte Carlo Image Synthesis
 Mathematics and Computers in Simulation
, 2000
"... A fundamental variance reduction technique for Monte Carlo integration in the framework of integroapproximation problems is presented. Using the method of dependent tests a successive hierarchical function approximation algorithm is developed, which captures discontinuities and exploits smoothness ..."
Abstract

Cited by 14 (0 self)
 Add to MetaCart
A fundamental variance reduction technique for Monte Carlo integration in the framework of integroapproximation problems is presented. Using the method of dependent tests a successive hierarchical function approximation algorithm is developed, which captures discontinuities and exploits smoothness in the target function. The general mathematical scheme and its highly efficient implementation are illustrated for image generation by ray tracing, yielding new and much faster image synthesis algorithms.
MonteCarlo Methods In Global Illumination
, 2000
"... lightsources, such as point or directional lightsources are preferred here, since their radiance is a Diracdelta like function, which simplifies the integral of equation (2.50) to a sum. These methods take into account only a single reflection of the light coming from the abstract lightsources. Ide ..."
Abstract

Cited by 11 (7 self)
 Add to MetaCart
lightsources, such as point or directional lightsources are preferred here, since their radiance is a Diracdelta like function, which simplifies the integral of equation (2.50) to a sum. These methods take into account only a single reflection of the light coming from the abstract lightsources. Ideal mirrors and refracting objects cannot be rendered with these methods. 2. Recursive raytracing Another alternative is to eliminate from the rendering equation those energy contributions which cause the difficulties, and thus give ourselves a simpler problem to solve. For example, if limited level, say n, coupling caused by ideal reflection and refraction were allowed, and we were to ignore the other nonideal components coming from nonabstract lightsources, then the number of surface points which would need to be evaluated to calculate a pixel color can be kept under control. Since the illumination formula contains two terms regarding the coherent components (reflective and refracting l...
Analysis of the quasimonte carlo integration of the rendering equation
 In Proceedings of Winter School of Computer Graphics
, 1999
"... szirmay�fsz.bme.hu Quasi�Monte Carlo integration is said to be better than Monte�Carlo integration since its error bound can be in the order of O�N ��1�� � � instead of the O�N �0�5 � probabilistic bound of classical Monte�Carlo integration if the integrand has �nite variation. However � since in c ..."
Abstract

Cited by 10 (3 self)
 Add to MetaCart
szirmay�fsz.bme.hu Quasi�Monte Carlo integration is said to be better than Monte�Carlo integration since its error bound can be in the order of O�N ��1�� � � instead of the O�N �0�5 � probabilistic bound of classical Monte�Carlo integration if the integrand has �nite variation. However � since in computer graphics the integrand of the rendering equation is usually discontinuous and thus has in�nite variation� the superiority of quasi�Monte Carlo integration has not been theoretically justi�ed. This paper examines the integration of discontinuous functions using both theoretical arguments and sim� ulations and explains what kind of improvements can be expected from the quasi�Monte Carlo techniques in computer graphics. Keywords � Rendering equation � quasi�Monte Carlo quadrature � Hardy�Krause variation 1
Stochastic transparency
 In I3D ’10: Proceedings of the 2010 ACM SIGGRAPH symposium on Interactive 3D Graphics and Games
, 2010
"... Stochastic transparency provides a unified approach to orderindependent transparency, antialiasing, and deep shadow maps. It augments screendoor transparency using a random subpixel stipple pattern, where each fragment of transparent geometry covers a random subset of pixel samples of size propor ..."
Abstract

Cited by 10 (1 self)
 Add to MetaCart
Stochastic transparency provides a unified approach to orderindependent transparency, antialiasing, and deep shadow maps. It augments screendoor transparency using a random subpixel stipple pattern, where each fragment of transparent geometry covers a random subset of pixel samples of size proportional to alpha. This results in correct alphablended 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 readmodifywrite loops other than traditional zbuffer 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.
Probability Trees
 IN GRAPHICS INTERFACE ’97
, 1997
"... A kD tree representation of probability distributions is generalized to support generation of samples from conditional distributions. An interpretation of the approach as a piecewise linear warping function is provided which permits a priori stratified sample generation. The representation is relat ..."
Abstract

Cited by 8 (0 self)
 Add to MetaCart
A kD tree representation of probability distributions is generalized to support generation of samples from conditional distributions. An interpretation of the approach as a piecewise linear warping function is provided which permits a priori stratified sample generation. The representation is related to higherorder spline estimators and representations via projection. An application in glyphbased volume visualization is presented.