Not logged in, Join Here! or Log In Below:  
 
News Articles Search    
 

 Home / 3D Theory & Graphics / Silhouette Account Manager
 
Archive Notice: This thread is old and no longer active. It is here for reference purposes. This thread was created on an older version of the flipcode forums, before the site closed in 2005. Please keep that in mind as you view this thread, as many of the topics and opinions may be outdated.
 
Andrew H. Cox

August 23, 1999, 08:51 PM

I noticed the algo for silhouette edge generation mentioned here and while the poly references edge structure (or winged-edge or half-edge or whatever it was)is fine for many things it is not necessary to detect silhouette edges.

For a set of polies, each with a list of vert indices simply go through each edge of the front facing polies and if the first vert of that edge is already stored in a temporary list that is being maintained (an array with one entry per vert with some sort of list hanging off each entry)at the index corresponding to the second vert it has been entered before from the other face that shares the edge and so the vert pair do not form a silhouette edge and the vert index just looked up can be deleted from the list it was just looked up in. If the first vert index wasn't found in the temp list 'hanging off' the second vert entry, the second index is added to the list 'hanging off' the entry corresponding to the first vert index.

At the end of that the temporary list contains for each vert index a list of the vert indices of the other ends of any silhouette edges incident on it.

I have this algo from a Brazilian academic by the name of Harlen Costa Batagelo.

I apologise for the confused tone of this post but I am in a hurry as I have to take some kids to Denmark in the morning. Feel free to mail me about this and I will get back to you with clarifications in a week.

BTW for anyone that knows me I haven't mailed because Outlook E. is crashing when I try to send: does anyone know what is up with it? I tried a re-install already.

Regards,

Andrew H. Cox

 
Jaap Suter

August 24, 1999, 06:55 AM

Hi,

this is a cool algo but...

I doupt whether this is fast actually. And since the winged edge structure never (almost) changes I see no probs with it.

But could you please explain whether your algo would be faster?

THanks,


Jaap Suter

 
Andrew H.Cox

August 24, 1999, 04:06 PM

Jaap Suter wrote:
>>Hi,
>>
>>this is a cool algo but...
>>
>>I doupt whether this is fast actually. And since the winged edge structure never (almost) changes I see no probs with it.
>>
>>But could you please explain whether your algo would be faster?

Hey,I just arrived in Denmark and the Youth Club I am staying in is a bloody palace, including a couple of online PCs in the entrance so my transmissions can continue :).

Extracting a vert sequence from a poly represented as an edge list needs a little bit of work per poly every time you want to draw it. My own approach to that is to cache a vert list for polies I wan't to draw to save redoing that work every frame. So why not go for the algo that provides the normal api's with the input they want and uses a common object format?

BTW it is Harlen's algo and he will soon be publishing a paper with results applying it to stenciled shadows.

Got to go now, more later.

Regards,

Andrew H. Cox

 
This thread contains 3 messages.
 
 
Hosting by Solid Eight Studios, maker of PhotoTangler Collage Maker.