Text archives Help
- From: Tim Holy <holy@wustl.edu>
- To: iv3d-users@sci.utah.edu
- Cc: Zhongsheng Guo <zsguo@pcg.wustl.edu>, Pei Sabrina Xu <xupei@wustl.edu>
- Subject: [IV3D-USERS] Improving NRRD/skewed volumes
- Date: Mon, 10 Oct 2011 14:01:07 -0500
Hi everyone,
After a long absence of not using IV3D, I am back to it.
I'm writing with a question and proposal. We have new data volumes in which
the coordinate axes, in physical space, are not orthogonal; the pixels are
therefore "skewed." By using the (new?) UVFReader tool, I stumbled across the
fact that UVF files appear to permit a generic transformation matrix to be
supplied. If so, this would seem to suggest that such volumes can be rendered
faithfully, as long as the proper transformation matrix is supplied.
First and foremost, is this true? I.e., is the support in Tuvok there for
this?
If so, then here's the proposal:
1. As far as I can tell, none of the "raw" importers can currently handle
volumes with non-orthogonal axes. However, the NRRD data format
(
http://teem.sourceforge.net/nrrd/format.html, version 4 onwards) permits the
necessary information to be supplied (see the "space directions" field). If
there is no better solution, I am offering to work on Tuvok/IO/NRRDConverter.
{h,cpp} to implement support for parsing the relevant fields.
2. I submit a patch; once it looks decent to you, we then change the syntax
of
the calls to ConvertToRAW away from
ConvertToRAW(...,FLOATVECTOR3& vVolumeAspect,...)
to something with an argument supporting generic transforms.
3. I patch the other converters (13 in total, including NRRD?) to accept this
different argument input. In those cases, it would simply fill the diagonal
of
the matrix, i.e., restoring the current behavior.
Of course, I welcome alternative suggestions.
Best,
--Tim
- [IV3D-USERS] Improving NRRD/skewed volumes, Tim Holy, 10/10/2011
Archive powered by MHonArc 2.6.16.