Based on IdeaStation requests we have been working on including polygonal meshes in Fusion 360. Our first steps are to create the new data model and supporting basics like move, applying materials and display characteristics. We plan on reading STL, OBJ as well as Alias meshes. Concurrently we are developing how the mesh can be used in the design process and for visualization. Thanks for the feedback on the IdeaStation. We will show you more as work progresses.
Can't tell you how important this development is for me. The real value in Fusion 360 will be to capture design intent by snapping tspline faces to selected points on the mesh. Allow me to give you a real world example. In one business I'm involved with we use
geometry from centuries old buildings and churches etc. throughout Europe and then replicate parts for buildings and houses in the US. Now I can just have my associates in Europe walk up to a piece of a castle or manor house that is lying in ruins in a field somewhere in Europe and scan it with a handheld scanner. Then I can take the mesh, which will be pretty rough because these ruins have been exposed to the elements for hundreds of years, and I can snap tsplines faces to selected mesh nodes and thereby recover
the design intent and skill of some master craftsmen from many centuries in the past. Then from my perfect Fusion 360 model I can
CNC the molds to make as many Louis the XIV fireplaces as I need. I could give you many more examples from Aerospace to
Bio Med. But you get the point.
Building over mesh data is one of the workflows we are targeting. In fact we are working with scans from ReCap which you might want to try if you haven't. ReCap is available as an online service now too. We are also investigating research being done to automatically create T-Spline data from scans.
If you are interested in scanning and mesh workflows you may want to check out Memento in the Autodesk Labs. It is a very simple program for fixing and decimating scan meshes.
Thanks for the info. So far all of the scan mesh manipulating software that I've tried works semi-ok for scans where the signal to noise ratio is high. They usually break down badly howeven when the SNR is low as it is in many cases. At that point it requires a person who is familiar with the design rules used by the original designers to discern which data is valid and which is noise. I've discussed this issue in depth with orthopedic surgeons who were required to reconstruct crushed bones in severe tramatic injury cases. They know what the bones should look like and they have MRI data showing them what they have to work with. The trick is to adequately model a high quality ideal surface off of a badly mangled existing data cloud. Tsplines seems to me to be an ideal tool because it requires very few data points to arrive at a high quality surface. Fusion 360 could be ideal in these cases because the interface is highly intuitive and the modeling capabilities are so good at both organic and mechanical modeling.
15" Macbook Pro Retina - OSX 10.9 in Portland, Oregon
Frank (@helmi / Twitter)
MacOSX 10.9 (iMac 2013)
Please support my ideas and comments to improve Fusion360. (link removed as it doesn't work anymore, see my ideas in the idea section)