Advertisement

Using Collision detection to determin visibility

Started by August 27, 2001 03:02 PM
6 comments, last by Frank Henry 23 years, 5 months ago
hi all. the idea behind it is to place a bounding region around the ''player'' and test for collisions, and if a collision occurs, draw the object. i have been thinking about this the whole day. what do you think about it?
Um, maybe this is a stupid question, but why would you want to do that ?
Advertisement
mostly to only draw the visible portions of a scene.
i''s just something i''ve been thinking of.
This is pretty standard, but its generally known as ''view frustum'' culling. Instead of creating another bounding box around the player, you use the coordinates of the view frustum (which you generally need anyway) and clip against them.

oh, cool!
thanks for the answer!
does anyone know of a exaple?
thanks
Advertisement
For Visibility you might want to look at an algorithm by D. T. Lee... Its a pretty simple algorithm that creates Visibility Polygon''s from a point in a plane...

The algorithm defined only works for polygons without holes, but it can be adapted to work for poly''s with holes...

I dunno where you can find specs about it online, but if you want I can mail you the pdf or something
quote:
Original post by Frank Henry
does anyone know of a exaple?



Theres a really good tutorial on view frustum culling at www.markmorley.com

[size="1"]

This topic is closed to new replies.

Advertisement