Einscan-S initial review

[24c]

einscan-files.jpg

This is what the files look like, before they are completed? You can see the original scan images increasing by 1, and it looks like they always start from this numbering scheme, and overwrite previous ones.

You're only looking at the 23/08/2015 12:38 to 12:47 date stamp. Looks like the ScanData_2rge.0000 is a place holder for the first aligned images and ScanData_9rge.0000 is the last one.
Is there a way to open these Alfred_T_S?

Mike

[Hugues]

einscan-files.jpg

This is what the files look like, before they are completed? You can see the original scans images increasing by 1, and it looks like they always start from this numbering scheme, and overwrite previous ones.

yes, i tried to open them with different software but no luck, probably a proprietary format of Shining3d. YOu can use good old notepad or texteditor under Mac to look at the header of those files, you see the column names. It's surely the scan data with information about normals or other stuff like this. I don't think we could extract much more info than what you get with the ascii files.

But Shining3d coud really use these files to recover from a crash, or continue a scanning session. I guess they know how to do it but maybe decided to give us a "dumbed down" version of their software for whatever marketing reason, as they sell higher end scanners and surely the corresponding software.

[24c]

But Shining3d coud really use these files to recover from a crash, or continue a scanning session.

I don't think it is dumbed down, just not enabled yet. It seems possible, after looking at the headers, they use part of the 3DScan v3.0 software found here
http://catalogue.fitman.atosresearch.eu/enablers/3dscan
I just glanced at the documentation, but it doesn't seem to make sense now. The images on the website look similar to some I've seen, but could be a dead end. :/
No doubt things will change, but I read there was a memory bug with the earlier version, so maybe that's why things are starting slowly. Also for the target audience, the scanner is a good fit in their eyes with the sister printer.

Mike

[24c]

This is where i'm struggling. At the end of this operation, i could still see some black regions on the scan,

Well Hugues, I got this far and the black bit, turned on the "light bulb", and the black object turned whiter. It's still black in places, but the Poisson filter absolutely blows the model up and distorts it beyond recognition. I can't export the mesh before this, and get anything usable into CAD, so I'll try another way, to get to the same point. I used bits of this tutorial before http://www.instructables.com/id/Usin...-Dat/?ALLSTEPS,
Thanks
Mike
PS I can use the .stl file, but I wanted to try using the raw scan data.

[Hugues]

This guy explains how to have Meshlab calculate the normals and why, starts to be interesting around 4 min,

https://www.youtube.com/watch?v=kVGENYi4gz8

I think my problem (and maybe yours) is due to the fact that i scanned a rather large model and the scanner does not remain in a fixed position but turns around the model in a vertical plane. So when you tell Meshlab to compute normals with regards to the viewpoint, the problem is there is only one viewpoint for the whole scan and it might not be where your scanner was.

One way to counter this i think is to load each point cloud individually and compute the normals independently for each scan and not for the collection of scans. It takes a bit more time but it probably gets rid of the black spots this way.

My hypothesis...

[24c]

OK Hugues, will give it a look, thanks for the input... just finished reading your 2 year EV bike thread, was LMAO sometimes, but respect congrats bud. 8)

[Hugues]

...

One way to counter this i think is to load each point cloud individually and compute the normals independently for each scan and not for the collection of scans. It takes a bit more time but it probably gets rid of the black spots this way.

My hypothesis...

Hypothesis confirmed

- So I loaded each scan into Meshlab, but not merged them right away
- Hide each scan but the first one, using the eye icon in the layers panel
- Display the axes in menu Renders / Show Axis
- By looking at the first scan and the axis, define what is the best camera XYZ position to see all the scan in a perpendicular manner
- Call in the menu Filters / point Set / Compute Normals for point sets /
- Select 100 for NEighbour / Chek flip normals
- Enter your camera position in the next 3 boxes in XYZ coordinate. YOu don't have to be precise. I select a point far away from the scan , normally 0, 0, 1000, or 0,-1000, 1000. Pay attention to negative side of axes.
- Click apply, and after a few seconds you see the result, make sure the light is turned on.
- If all ok, don't close this dialog, select the next scan in the layers panel, make it visible, check if your camera position needs to be updated, then click apply
- repeat until last scan,

then merge all scans and create your mesh with the Poisson filter.

One more week on holidays, kids back to school, so plenty of time to explore.

[Harry Cayne]

I agree with you Nerv the water sealing should be optional and we should be able to set the number of images the scan file is made, depending of the complexity of the object. I rather wait 30 minutes for a detailed but small object than to use it in manual mode. Thanks for your feedback.
Harry

[ekmek]

Great info in this thread. I've been looking at 3D scanning but haven't bought one. I mainly want it to do it so me and my son can make custom 4" action figures. I'm pretty good with Blender 3D so I figure I can handle edits of a scanned model.

So my questions: EinScan-S can it due a detailed enough resolution for an action figure head (they are about half inch diameter)?

How much does this cost roughly?

If I mail an action figure head is anyone willing to scan it so I can take a look at the quality?

thanks!

[Hugues]

Great info in this thread. I've been looking at 3D scanning but haven't bought one. I mainly want it to do it so me and my son can make custom 4" action figures. I'm pretty good with Blender 3D so I figure I can handle edits of a scanned model.

So my questions: EinScan-S can it due a detailed enough resolution for an action figure head (they are about half inch diameter)?

How much does this cost roughly?

If I mail an action figure head is anyone willing to scan it so I can take a look at the quality?

thanks!

I think it should be good enough, but really depends on the size of the features you are trying to scan. Are you going to 3d print them ? If you are using a FDM printer, then the scanning accuracy is more than enough.

You can send me an example of your action figure if you want me to scan it for you, but i live in Switzerland, you might try to find someone in the USA. Just send me a private message through the forum if you want my address.

As for the price:
http://www.einscan.com/#!shop/cwcm