PDA

View Full Version : Shadow map sampling techniques


pgregory
08-25-2003, 09:05 PM
Does anyone here have any advice/links/references to information about efficient shadow map filtering techniques?

I have implemented the classic "percentage closer filtering" in the Aqsis renderer, but wondered if there was a better/more efficient method. Preferably one that lends itself well to real/faked soft shadows.

Thanks for any help.

PaulG

gga
08-26-2003, 03:39 AM
For filtering itself, check:

Efficient Shadow Antialiasing Using an A-Buffer
Brett Keating
Lawrence Livermore National Laboratory
Livermore, California


For scanline improvements, albeit not entirely related to just filtering, you can check:

Shadow mapping based on dual depth layers ( a better mid-point shadow map approach )
http://www.vis.uni-stuttgart.de/eng/research/pub/pub2003/eg03-weiskopf.pdf

Adaptive shadow maps
http://www.graphics.cornell.edu/pubs/2001/FFBG01.pdf

Perspective shadow maps
Siggraph 2002

Second Depth Shadows
http://www.cs.unc.edu/~lastra/Courses/Papers/Wang_Molnar_Second_Depth_Shadow.pdf

For linear lights
http://www.cs.ubc.ca/~heidrich/Papers/RW.00_1.pdf


For more real-time and less accurate approaches, check out:

http://w3imagis.imag.fr/Publications/2003/HLHS03/RTSoftShadow
s-lowRes.pdf

which is a nice overview of the different approaches.



With raytracing, there's the obvious monte carlo sampling of an area light.
Other cheaper approaches I know with raycasting/tracing are:

Cone Shadows
http://www.cs.utah.edu/~bes/papers/coneShadow/shadow.html

pgregory
08-27-2003, 02:07 PM
Thanks for those references. The A-Buffer one from Brett Keating, seems to be the most useful for my requirements.

They will all make interesting reasing though.

Cheers

PaulG

richk
08-28-2003, 07:52 PM
That Brett Keating paper sounds interesting - don't suppose you have an electronic copy I could nick? Doesn't seem to be available online, and my copies of JGT don't go back anything like that far...

Or could you give a quick description of how it works - the abstract on the JGT site is a little too, erm, abstract to guess many of the details!

gga
08-29-2003, 05:34 AM
Originally posted by richk
That Brett Keating paper sounds interesting - don't suppose you have an electronic copy I could nick? Doesn't seem to be available online, and my copies of JGT don't go back anything like that far...



Sorry, no. Don't know how it works in the UK, but in the States, you can visit some prestigeous public college library to obtain copies of old papers or journals.


If not, contact the publisher directly.

From their website (hard to find but info is there)...

Back volumes, single issues, and reprints

journal of graphics tools
A K Peters, Ltd.
63 South Avenue
Natick, MA 01760-4626
Phone (508) 655-9933, Fax (508) 655-5847
service@akpeters.com

For price and availability information, please contact A K Peters, Ltd. as listed above.

Mauritius
09-03-2003, 12:19 PM
Just for completeness: Andrew Chapman just posted the "Perspective Shadow Maps" paper's link (http://www-sop.inria.fr/reves/publications/data/2002/SD02/PerspectiveShadowMaps.pdf) to c.g.r.r.
This guy also said

CGTalk Moderation
01-15-2006, 11:00 PM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.