research about shadow map

[rendermaniac]

A shadow map is a depth map from the point of view of the camera. Usually it also stores the transform information of where it was generated. This is so that in the main render for each point being shaded you transform it into the space of the shadow map (which is why you store it in the shadow map) which gives you the shading point position as if the light was the origin.

You then compare this to the depth in the map at that point (also in the light coordinate system). If the depth in the map is closer to the light than the point you are shading then it is in shadow otherwise it is lit by that light.

In fact it is basically a way of saving out a ray trace ray - and people have used it in this way - search for the 2003 RenderMan Siggraph coursenotes on http://www.renderman.org/ by Christophe Hery. That code does almost the same thing as a shadowmap function - he just uses it to find the shadowing point rather than comparing it to see if it's in shadow.

There are other types as well such as midpoint shadows and soft shadows.

I would highly recommend looking at RenderMan to learn about this. The RIB format is very simple any shows you all the details. The RenderMan Companion has the code for a program to place a shadow camera properly. Also check out Advanced RenderMan which is more up to date and tells you pretty much everything you need to know. Also Computer Graphics: Principles and Practise goes through pretty much every CG technique in detail, but is harder to read. And with RenderMan there are free renderers available such as Aqsis http://www.aqsis.com/ and 3delight http://www.3delight.com/ which even has a free installer http://www.plastickitten.net/liquidwiki/index.php/Main_Page (both work with Liquid so you can output from Maya if you have it).

If you want to be really fancy you could try implementing something like perspective shadow maps. That may be a little tricky though - would love to know if you manage that one!

Hope that helps

Simon

[Andrew W]

A shadow map is a depth map from the point of view of the camera.
where the "camera" is the light position, i.e. as if you were "looking through" the light and where the frustrum is defined by the light's coneangle.

A

[CGTalk Moderation]

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.