[SML] Canvas 3D modeling software

[Steven Hood]

There are "point-cloud" mapping systems in use. I think Vector Works had an 
article about it on their website. I also remember seeing a show on the 
History Channel that used one to map a room with a staircase at the end (in 
a castle, IIRC) that they were studying.
Hth,
Steven R Hood


On December 5, 2016 6:10:51 AM Joe D via Stagecraft 
<stagecraft at theatrical.net> wrote:

> The company I work for installs large industrial refrigeration chambers 
> (think room, up to warehouse sized freezers).  I would love to be able to 
> do a simple scan of an existing facility.  But, when I spoke to some 
> vendors about the type of sensor used by the Canvas program, they said that 
> it is not nearly accurate enough.  It is not so much that an individual 
> scan is inaccurate, but that as you try to "paint" a large area, the errors 
> accumulate as each scan is added to the prior one.  While there is some 
> effort to reconcile the final scan to the first, I was advised that the 
> results are not useful at the scale of a warehouse.
>
> And for us, another issue is that we want to capture smaller profile 
> objects, such as the roofing trusses.  I don't need every detail, but this 
> type of scanner would not necessarily show the existence of the roof 
> trusses at all !
>
> To be able to "see" a roof truss, I think I would need a laser line 
> generator type of scanner.  I don't think such hardware should be 
> expensive.  But, I have never seen anything like this, designed for large 
> room and warehouse scale scans, at any price.
>
> While my earlier description of a "Scan your paint-Choose a color-Generate 
> a mix formula" software idea has not generated any interested yet, I will 
> go ahead and propose my laser scanner idea, in case someone picks it up, 
> and makes what I want!
>
> There exist these types of laser scanners for small scale objects. It is 
> typically a camera mounted together with a laser line generator.  Sometimes 
> the object rotates in front of it.
>
> I propose that the camera and laser sit on a tripod base, and this 
> arrangement rotates around.  It may be able to obtain 5 or 10 degrees 
> vertically of data. As each rotation is complete, it shifts its aim a bit 
> higher.  One complication is that the laser and camera must be separated, 
> and aimed somewhat "cross-eyed". This means that the scanner has a specific 
> range. The further the two are separated, the more accurate the scan 
> becomes. This separation needs to be increased as the scanned object is 
> further away. Likewise, the zoom of the camera lens may also need to be 
> changed. The zoom lens is probably a bigger expense, but lasers are cheap. 
> So, it might just have multiple lasers at different distances.  A Red and 
> Green color scan could be done at once.  The user might manually change the 
> camera lens, and perhaps move lasers to make coarse changes to the range as 
> well.
>
> -Joe
>
>
> ____________________________________________________________
> For list information see <http://stagecraft.theprices.net/>
> Stagecraft mailing list
> Stagecraft at theatrical.net
> http://theatrical.net/mailman/listinfo/stagecraft_theatrical.net