Wayne Morellini |
November 26th, 2005 11:53 PM |
Quote:
Originally Posted by Larry Edwards
I'm not totally clear on what Wayne is proposing but depth/shape from focus/defocus has been a subject of research and [/url]
|
Haven't had access to that research. Yes, pretty much the sort of thing you said, that defocus information gives away point and limited 3D surround shape information, and because defocus size grows the further away you are from the in focus region you can get distance information from the it's size. One idea, I had probably in 97, was an optical system that acted as a lens, that could cover a whole set (I should say "small set", costly) and give a very good surround image, but that is tough to do, I don't remember if I ever got around to completely cracking it. But the interesting thing is, as posted elsewhere, is that I saw a Popular Science magazine in the newsagent yesterday, that had an article on a 4 Giga pixel hand made camera, that would go a long way towards solving the image resolution acquisition specs.
Quote:
Actually there is a big database of such info, old SIGGRAPH proceedings ;)
|
I meant more in general, for all ideas for all knowledge, an alternative repository to the patent database for an alternative system to patents.
I recently found out from a researcher friend, that people had beat me to the viewpoint interpolation idea in the 90's, and that is the only reason I mentioned it, already been done. I am working on something much better with 3D pixels (a bit similar to voxel). But the interpolation idea was just an idea to do something like the primitive 3D from 2D bit-mapped graphics of the 80's and early 90's, much better. That time is gone now, so it doesn't matter. But it would have required relatively simple hardware to implement compared to conventional 3D graphics. Who knows, we might still have been using 2D graphics for some 3D games, if such a thing had been implemented. There are many such lost opportunities in home computing. For example a few years ago I figured out a low memory vector to raster conversion method that would have been excellent if implemented in hardware on Atari VCS back in the 70's. If somebody had figured to implemented it in hardware back then, games, and the overall functionality of the VCS, would have been 10 times better (for anybody that doesn't know, computer memories were so small, and speeds so low, they produced very crude graphics that left little memory space left over for real functionality).
|