This thread is a good one. We should keep it up. I would welcome talking with people over the phone or some chat system--or we could keep it up right here on the mailing list.
Anyway- I read the Scillion paper quickly and it looks like it may be the ticket for my virtual MSU project. But again, the question is how to do the whole thing efficiently in Java 3D. And we should consider other related methods as well. I can probably throw some student help at the project + some limited funds at solving this problem. There seem to be two basic situations- generating scenes of real places on a semi-procedural basis and the generating of totally artificial worlds on a totally procedural basis. Maybe we should get a list together of those of you interested in this area. The document that David started with is a good one. We should start pasting in the article references below each category. On the virtual MSU, I have access to a map of the terrain that is in a .DWG file. Some of you tried to take a crack at it but is was too humongo to deal with. This file contains data about trees, etc. Now if I have it correctly, we need to have digital pictures of the distant scene elements. In navigating MSU, we start at the central element, Beuamont tower. It sits on a little hill, has a side walk, some trees, etc. I want to load up that terrain point and all of the 3D elements around it--perhaps even the trees--, put Beaumont on top and look around. The grassy area around Beaumont extends 200 yards with buildings starting there. I am assuming that we put our first imposters there. When we get close to the building, through LOD we start loading up a different part of the terrain/content model. BTW, remember Julian Gomez's idea of the scene graph as a relational database? Really, we are talking about creating a specialized data structure- basically a relational one. OK, I have to go back to data analysis. It does seem like we would benefit from working together on some of this stuff. Maybe David's rant has started something. Alex -----Original Message----- From: Discussion list for Java 3D API [mailto:[EMAIL PROTECTED]]On Behalf Of Raj Vaidya Sent: Saturday, January 05, 2002 2:51 AM To: [EMAIL PROTECTED] Subject: Re: [JAVA3D] Re-inventing the wheel Hi EveryOne: Thanks David Yazel for your article, and Best Wishes in your endeavor. Here are some links which you, and others on this List, might find useful: 1. http://citeseer.nj.nec.com/cs An excellent repository which has many papers on graphics including texturing, data structures, spatial partitioning etc. Specifically, you may find the following paper interesting: F. X. Sillion, G. Drettakis, and B. Bodelet. Efficient impostor manipulation for real-time visualization of urban scenery. Comp. Graphics Forum, 16(3), September 1997. 2. Siggraph proceedings: Surprising many of the papers are available online - collected by people to whom I owe many thanks. Do a search on google using the keywords: siggraph papers. Also some Siggraph Course Notes on Advanced Techniques (SGI site), subdivision surfaces (IBM site) are available. 3. EVLIB: The Electronic Visualization Library Excellent set of references including paper links, bibtext entries etc. Good Luck Raj Vaidya =========================================================================== To unsubscribe, send email to [EMAIL PROTECTED] and include in the body of the message "signoff JAVA3D-INTEREST". For general help, send email to [EMAIL PROTECTED] and include in the body of the message "help". =========================================================================== To unsubscribe, send email to [EMAIL PROTECTED] and include in the body of the message "signoff JAVA3D-INTEREST". For general help, send email to [EMAIL PROTECTED] and include in the body of the message "help".
