PDA

View Full Version : Level set and Fast Marching methods


Aikiman
11-14-2006, 07:51 PM
This is more of a low level programming method I've just been reading about...

http://en.wikipedia.org/wiki/Level_set_method

http://math.berkeley.edu/~sethian/

and just wondering if anyone out there knows much about this and whether it is implemented in Maya and if so where. These methods are being used for scientific research and medical practice like robotics and x-ray for example but I dont know if they use it in Maya. My guess is that currently no it isn't but that future development will incorporate some of the processes involved with level set methods. Does anybody have any idea at this stage?

sparaig
11-14-2006, 09:34 PM
This is more of a low level programming method I've just been reading about...

http://en.wikipedia.org/wiki/Level_set_method

http://math.berkeley.edu/~sethian/

and just wondering if anyone out there knows much about this and whether it is implemented in Maya and if so where. These methods are being used for scientific research and medical practice like robotics and x-ray for example but I dont know if they use it in Maya. My guess is that currently no it isn't but that future development will incorporate some of the processes involved with level set methods. Does anybody have any idea at this stage?

I'd guess that PhotoShop uses some of them, and many medical imaging packages certainly do, but I doubt if a generic 3D modeling and animation package like Maya or any of its relatives does anything like that. Maybe as plugins/addons.

Aikiman
11-14-2006, 11:39 PM
Thats the first thing I thought of sparaig, that photoshop uses the Fast marching method in the wand tool :). Although that seems to use a tolerance gauge on pixel contrast values only and not necessarily on the lumenance falloff rate, so to speak. (kind of like z-depth)

sparaig
11-15-2006, 07:57 AM
Thats the first thing I thought of sparaig, that photoshop uses the Fast marching method in the wand tool :). Although that seems to use a tolerance gauge on pixel contrast values only and not necessarily on the lumenance falloff rate, so to speak. (kind of like z-depth)

Yeah. I have a real interest in this stuff right now since I'm trying to model a human brain based on the harvard brain atlas MRI images, which apparently aren't available any more

http://www.med.harvard.edu/AANLIB/_README_


I'm tempted to implement the marching cubes algorithm and work with the output, but from what I've seen of computer-generated stuff is that it either costs thousands of dollars or is too cruddy-looking to be useful (realistic brains are UGLY).

djwarder
11-21-2006, 05:52 PM
Hmm, I don't think those methods are used in Maya, but they're definately relevant to high-end, state-of-the-art fluid sims (i.e Fedkiw's stuff, but not Realflow!) ...

CGTalk Moderation
11-21-2006, 05:52 PM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.