search

FNNDSC / projectman

FNNDSC project manager
3 stars 0 forks source link

write description for the patent #66

Open NicolasRannou opened 9 years ago

NicolasRannou commented 9 years ago

I do not put details here just in case... :)

But a few details which detail uniqueness/novelty of the approach is all we need. Maybe some graphics too. Will be used in the paper too anyway so worth the effort.

NicolasRannou commented 9 years ago

https://docs.google.com/a/babymri.org/document/d/1gu9Y1J-7i3pcVsyDILFv5Q4iVT4hA7Nsy4v_AejKSos/edit#

Waiting for devs answers to know if this is enaough to get started with before closing.

rudolphpienaar commented 9 years ago

Looks good for a start!

How sure are we about the current way of doing things? It seems to me that anyone who knows about GPU Web/Open/GL(ES) would think to do it like this, or no?

NicolasRannou commented 9 years ago

Well right, in the context of slices, the 2D texture is always generated before being pushed to the graphic card. The only time calculations are happening on the graphic card is for volume rendering (VTK, XTK, etc.). Our approach basically is to do a volume rendering of the 2D slice. So nothing really new but that is a different approach than what people are currently doing to display a 2D slice.

NicolasRannou commented 9 years ago

I added some extra information - should it be more detailed to send ti to TIDO? @rudolphpienaar