Publications | Pascal Grittmann

Reviewing conflicts based on recent co-authors (as of 2024): Philipp Slusallek, Ömercan Yazici, Iliyan Georgiev, Alexander Rath, Sebastian Herholz, Philippe Weier, Qingqin Hua, Miša Korać, Corentin Salaün, Karol Myszkowski, Gurprit Singh

Peer-reviewed journal papers

Revisiting correlated mixture sampling for rendering applications

Qingqin Hua, Pascal Grittmann, Philipp Slusallek

ACM Trans. Graph. (SIGGRAPH 2023) 10.1145/3592435

Revisits and generalizes the math behind optimal multiple importance sampling to make it more practical

pdf GitHub bibtex

Efficiency-aware multiple importance sampling for bidirectional rendering algorithms

Pascal Grittmann, Ömercan Yazici, Iliyan Georgiev, Philipp Slusallek

ACM Trans. Graph. (SIGGRAPH 2022) 10.1145/3528223.3530126

Shows how to optimize parameters for bidirectional rendering algorithms (e.g., number of light paths, number of connections).

pdf GitHub bibtex

EARS: Efficiency-Aware Russian Roulette and Splitting

Alexander Rath, Pascal Grittmann, Sebastian Herholz, Philippe Weier, Philipp Slusallek

ACM Trans. Graph. (SIGGRAPH 2022) 10.1145/3528223.3530168

Allocates computation where it matters most by optimizing path termination probabilities and splitting factors.

pdf GitHub bibtex

Correlation-aware multiple importance sampling for bidirectional rendering algorithms

Pascal Grittmann, Iliyan Georgiev, Philipp Slusallek

Computer Graphics Forum (Eurographics 2021) 10.1111/cgf.142628

Identifies and fixes an MIS weighting issue in bidirectional rendering algorithms. While not theoretically optimal, the fix is practical and easy to compute.

pdf GitHub bibtex

Variance-Aware Path Guiding

Alexander Rath, Pascal Grittmann, Sebastian Herholz, Petr Vévoda, Philipp Slusallek, Jaroslav Křivánek

ACM Trans. Graph. (SIGGRAPH 2020) 10.1145/3386569.3392441

Derives a better target function that path guiding methods should learn - faster rendering for free!

pdf GitHub bibtex

Variance-aware multiple importance sampling

Pascal Grittmann, Iliyan Georgiev, Philipp Slusallek, Jaroslav Křivánek

ACM Trans. Graph. (SIGGRAPH Asia 2019) 10.1145/3355089.3356515

Identifies and fixes an MIS weighting issue in bidirectional rendering algorithms. The fix − injecting variance estimates − is generally applicable to any MIS estimator.

pdf GitHub bibtex

Optimal Multiple Importance Sampling

Ivo Kondapaneni, Petr Vévoda, Pascal Grittmann, Tomáš Skřivan, Philipp Slusallek, Jaroslav Křivánek

ACM Trans. Graph. (SIGGRAPH 2019) 10.1145/3306346.3323009

Derives the theoretically optimal weights to use when combining multiple sampling techniques, and shows that these weights can be computable in some simple practical applications.

pdf bibtex