[x3d-public] ideal points in IndexedFaceSet

Andrey Filippov andrey at elphel.com
Fri Feb 16 15:39:56 PST 2018


This is why I though that boolean vertex attribute may work better. We will try your suggestions with shaders, what I meant was to get rid of Background node and make its functionality more uniform with other nodes.

The faces that have all vertices at infinity will not get any light from a point source, and there will be no normal point. But a triangle with only one infinite point can be a road going to the horizon. With two infinite points - a segment between two rays from the common point. Both can get correct light from the point source and have the normal point. Such objects are mathematically very simple and intuitive. Implementation - I do not have enough experience, maybe it is possible just to parse such objects and replace faces with the easier to render primitives?

Allowing these infinite points (directions) may have practical applications for autonomous vehicles, for example - it is not just our case when some objects are too far to have a known distance value, but they should otherwise (where possible) behave almost as normal (near) objects. And they can become near ones if the vehicle goes in their direction. It would be a cleaner solution than putting such objects at large, but finite distance, but such replacement can be performed when converting to other 3-d representations that do not have such node.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://web3d.org/pipermail/x3d-public_web3d.org/attachments/20180216/e5fe6faf/attachment.html>

More information about the x3d-public mailing list