Object Recognition in SL

Hello

I would like to know if there is any work being done on object Recognition in SL. Using Polygon not prims? is that possible?. it would be a way for an Intelligent avatar or a human controlled one to collect information in a different way or it could be use for tracking as well rather than using the method currently used. In some 3D simulation engine they use different visual system either pixels, voxels (color) or polygons. Will SL support any of those feature or is there a work around rather than using unique ID and coordinate.

Thanks for info

Ousava

Interesting question. I can see a couple of ways one might proceed, but I don't think there is anything built in.

1) You could build an engine that scrapes the pixels off the viewer screen, and feeds that into your vision tools.

2) You could use libSecondLife (or whatever it is called) to get the graphics info, and render it straight into your vision tools.

3) You could modify the Open Source viewer yourself to add hooks to your vision tools.

4) You could intercept all the OpenGL calls, and route them to your vision tools.

In SL, almost of all of the visual data is downloaded as objects and their positions in space. Almost all the rendering is done in the viewer on your desk top.

Note that I do not really know what I am talking about, but I think that's the ways you could proceed.

thanks for info

It is something i know that other people are working on Object Recognition in SL