| |
Hi Darrel,
> Here are the STL file for the Anakin Skywalker hair piece and the Senate
> Commando helmet.
>
> I’m not really satisfied with how the front of the part looks; if you
> look at the picture on Peeron there’s a lot more definition in the hair
> (it’s very sharp-edged, like the anime-hair pieces) but the algorithm in
> the meshing command that combines multiple scans (in this application
> called the “fuser”) smoothed it out too much. I think part of what
> caused that is I left the head of the minifig on during the scan
> process. I thought that would be a good reference point later for people
> trying to scale up the model to LDU, but I think it added unnecessary
> data to the file (you can see it added filleted edges where the head and
> the hair intersect, when obviously there is no curve there).
Indeed, the face is not really needed here. Your scanner properly creates a file
in millimeter units, scaling by 2.5 directly provides LDU.
>
> It is also possible to save a version of the file without the fusing
> applied (this really looks messy as it is a combination of – in these
> cases – 8 different scans revolved around 360 degrees). If you would
> rather have the un-fused version let me know.
From my experience with laser scanner, some smoothing can't be avoided because
the part itself is not perfect, and there is some rounding during scanning and
more during fusing. After searching a bit, I found a little software gem,
Sculptris, that allows to enhance mesh before down-scaling it. See the full
story here:
http://www.david-laserscanner.com/forum/viewtopic.php?f=2&t=1972&p=12247
>
> I can try again without the head in place to give a harder edge to the
> scan and maybe get more definition to the hair itself. I can also try a
> multitude of different settings in the scanner – I’m just learning how
> to use it and I’m sure there are some tweaks that could help with the
> results. I’m not sure how much detail is really needed for the DAT file
> to convey the shape.
Present version is too detailed for a LDraw part, but it's easier to start with
a fine mesh (finer than this one) to use Sculptris. After enhancement, I
downsample the mesh using Meshlab.
> I suppose if anything, the digital scan could be a good reference to
> measure these kinds of parts (instead of having to use calipers and such).
>
> Let me know if you need anything else or if you want me to try any
> different settings.
One thing is missing: bottom of scan (especially in the hair). Generally I use
16 scans, 8 around and above part, and 8 below it to make sure to get all data.
Philo
|
|
Message has 1 Reply:
 | | Re: 3D laser scanning
|
| (...) Thanks, Philo! In that case, I shall rescan without the head, from more angles, and with finer settings - I was trying (unsuccessfully) to try and limit the polygon count by scanning in lower resolution, but if you're using other software to (...) (14 years ago, 23-Mar-11, to lugnet.cad, lugnet.cad.dat.parts)
|
Message is in Reply To:
| | Re: 3D laser scanning
|
| Hi Philo, I've sent you the STL file for the Anakin Skywalker hair piece and the Senate Commando helmet. I’m not really satisfied with how the front of the part looks; if you look at the picture on Peeron there’s a lot more definition in the hair (...) (14 years ago, 22-Mar-11, to lugnet.cad, lugnet.cad.dat.parts)
|
31 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
