1. [Publications](/index.php/publications)
2. Massively Parallel Path Space Filtering
 
 # Massively Parallel Path Space Filtering

  ![](/sites/default/files/styles/wide/public/pubs/2019-08_Massively-Parallel-Path/default.jpg?itok=ePHhQZsa)

 Restricting path tracing to a small number of paths per pixel for performance reasons rarely achieves a satisfactory image quality for scenes of interest. However, path space filtering may dramatically improve the visual quality by sharing information across vertices of paths classified as proximate. Unlike screen space-based approaches, these paths neither need to be present on the screen, nor is filtering restricted to the first intersection with the scene. While searching proximate vertices had been more expensive than filtering in screen space, we greatly improve over this performance penalty by storing and looking up the required information in a hash table. The keys are constructed from jittered and quantized information, such that only a single query very likely replaces costly neighborhood searches. A massively parallel implementation of the algorithm is demonstrated on a GPU.



 ## Authors



[Nikolaus Binder](/index.php/person/nikolaus-binder)

Sascha Fricke (University of Braunschweig)

[Alex Keller](/index.php/person/alex-keller)

 

 

 ## Publication Date



Friday, February 15, 2019

 

 ## Published in



[arXiv](https://arxiv.org/abs/1902.05942?context=cs)

 

 ## Research Area



[Algorithms and Numerical Methods](/index.php/research-area/algorithms)

[Computer Graphics](/index.php/research-area/computer-graphics)

[High Performance Computing](/index.php/research-area/high-performance-computing)

[Real-Time Rendering](/index.php/research-area/real-time-rendering)