ACM Home Page
Please provide us with feedback. Feedback
Reordering for cache conscious photon mapping
Full text PdfPdf (257 KB)
Source GI; Vol. 112 archive
Proceedings of Graphics Interface 2005 table of contents
Victoria, British Columbia
SESSION: Rendering table of contents
Pages: 97 - 104  
Year of Publication: 2005
ISBN ~ ISSN:0713-5424 , 1-56881-265-5
Authors
Joshua Steinhurst  University of North Carolina at Chapel Hill
Greg Coombe  University of North Carolina at Chapel Hill
Anselmo Lastra  University of North Carolina at Chapel Hill
Sponsor
CHCCS : The Canadian Human-Computer Communications Society
Publisher
Canadian Human-Computer Communications Society  School of Computer Science, University of Waterloo, Waterloo, Ontario, Canada
Bibliometrics
Downloads (6 Weeks): 5,   Downloads (12 Months): 34,   Citation Count: 2
Additional Information:

abstract   references   cited by   index terms   collaborative colleagues  

Tools and Actions: Review this Article  

ABSTRACT

Photon mapping is a global illumination algorithm for generating and visualizing a sparse representation of the incident radiance on surfaces. Photon mapping places an enormous burden on the memory hierarchy. A 512x512 image using the standard kd-tree data structure requires more than 196GB of raw bandwidth to access the photon map. This bandwidth is a major obstacle to our long term goal of designing hardware capable of real time photon mapping.This paper investigates two approaches for reducing the required bandwidth: 1) reordering the kNN searches; and 2) cache conscious data structures. Using a Hilbert curve reordering, we demonstrate an approximate lower bound of 15MB of bandwidth. This improvement of four orders of magnitude requires a prohibitive amount of intermediate storage. We then demonstrate two more cost-effective algorithms that reduce the bandwidth by one order of magnitude to 24GB with IMB of storage. We explain why the choice of data structure can not, by itself, achieve this reduction. Irradiance caching, a popular technique that reduces the number of required kNN searches, receives the same proportional benefit as the higher quality photon gathers.


REFERENCES

Note: OCR errors may be found in this Reference List extracted from the full text article. ACM has opted to expose the complete List rather than only correct and linked references.

 
1
R. Bayer and E. McCreight. Organization and maintenance of large ordered indexes. Acta Informatica, 1(3):173--189, 1972.
2
 
3
 
4
Per H. Christensen and Dana Batali. An irradiance atlas for global illumination in complex production scenes. In Proc. Eurographics Symposium on Rendering, pages 133--141, 2004.
5
6
 
7
 
8
V. Havran. Analysis of cache sensitive representations for binary space partitioning trees. Informatica, 23(3):203--210, May 2000.
 
9
10
 
11
Intel. Microburst Architecture. IA-32 Intel Architecture Software Developers Manual, 1:37--40, 2003.
 
12
Wojciech Jarosz, April 2004. http://renderedrealities.net/.
 
13
 
14
Yuanxin Liu and Jack Snoeyink. A notation for hilbert curves to support multidimensional spatial indexing. In preparation.
 
15
16
 
17
 
18
NVIDIA. Ultra-high-end products, October 2004. http://nvidia.com/page/qfx_uhe.html.
 
19
 
20
 
21
 
22
 
23
E Reinhard and E W. Jansen. Rendering large scenes using parallel ray tracing. In First Eurographics Workshop on Parallel Graphics and Visualization, pages 67--80, 1996.
24
 
25
 
26
Ingo Wald, Johannes Guenther, and Philipp Slusallek. Balancing considered harmful - faster photon mapping using the voxel volume heuristic. In Proc. Eurographics, 2004.
 
27
Ingo Wald, Philipp Slusallek, Carsten Benthin, and Markus Wagner. Interactive rendering with coherent ray tracing. Computer Graphics Forum, 20(3):153--164, 2001.
28


Collaborative Colleagues:
Joshua Steinhurst: colleagues
Greg Coombe: colleagues
Anselmo Lastra: colleagues