Technical Papers

Sampling & Spectra

Tuesday, 12 August 9:00 AM - 10:30 AM | Vancouver Convention Centre, East Building, Ballroom A Session Chair: Steve Marschner, Cornell University

A Constructive Theory of Sampling for Image Synthesis Using Reproducing Kernel Bases

An alternative theory of sampling for image synthesis that does not rely on Dirac deltas but instead employs reproducing kernels for the correspondence between pointwise samples and a continuous function. Practically, the theory enables pointwise numerical techniques designed for light transport that provide significantly improved performance.

Christian Lessig
University of Toronto

Mathieu Desbrun
California Institute of Technology

Eugene Fiume
University of Toronto

Fast Tile-Based Adaptive Sampling With User-Specified Fourier Spectra

Introducing a novel tile-based method for adaptive two-dimensional sampling. The approach achieves several orders of magnitude speed improvement over current spectrum-controlled sampling methods. The result is a linear-time, adaptive, and high-quality sampling of arbitrary density functions that conforms to the desired spectral distribution.

Florent Wachtel
Université Claude Bernard Lyon 1

Adrien Pilleboue
Université Claude Bernard Lyon 1

David Coeurjolly
CNRS/LIRIS

Katherine Breeden
Stanford University

Gurprit Singh
Université Claude Bernard Lyon 1

Gael Cathelin
Université Claude Bernard Lyon 1

Fernando de Goes
California Institute of Technology

Mathieu Desbrun
California Institute of Technology

Victor Ostromoukhov
Université Claude Bernard Lyon 1, CNRS/LIRIS

A Local Frequency Analysis of Light Scattering and Absorption

A novel analysis of absorption and scattering of local light fields in the Fourier domain is combined with covariance tracing to enable efficient estimation of frequency content; this approach improves various rendering algorithms of participating media.

Laurent Belcour
Université de Montréal

Kavita Bala
Cornell University

Cyril Soler
INRIA Rhône-Alpes

Boosting Monte-Carlo Rendering by Ray Histogram Fusion

This paper proposes a new multi-scale filter for acceleration of Monte Carlo renderers. Each pixel is characterized by the colors of the rays that reach its surface. The algorithm is universal. It produces a significant gain of PSNR. And it is directly extendable to video.

Mauricio Delbracio
École normale supérieure de Cachan, Duke University

Pablo Musé
Universidad de la República

Antoni Buades
École normale supérieure de Cachan, Universitat de les Illes Balears

Julien Chauvier
e-on software, inc.

Nicholas Phelps
e-on software, inc.

Jean-Michel Morel
École normale supérieure de Cachan

Factored Axis-Aligned Filtering for Rendering Multiple Distribution Effects

A method to adaptively sample and filter for simultaneous rendering of defocus blur, soft shadows, and indirect illumination. The approach proposes factoring radiance into texture and irradiance, which allows for pre-filtering irradiance, and uses different sampling rates for primary and secondary effects, further reducing the overall ray count.

Soham Uday Mehta
University of California, Berkeley

JiaXian Yao
University of California, Berkeley

Ravi Ramamoorthi
University of California, Berkeley

Frédo Durand
CSAIL MIT