Would it be possible for a beta user to scan some bones (if they have access to such materials)? Or perhaps do a scan of a large, feature rich object that reaches the maximum size capacity of the MINI? I understand the MINI is more tailored to smaller items but I see the official specs show a recommended maximum size of 50 x 50 x 50 cm.
I plan on using this scanner in a museum setting to scan larger animal skulls for research.
If I have a large item, would it be possible to make multiple scans and join them together? For example, the maximum size would be around 90 x 70 x 70cm. Or would that be a huge strain on the computer with very large file sizes?
With the smaller scanning area of the MINI, would the larger item take an unreasonably long time to scan?
3.The intended items are feature rich, would tracking be an issue for larger objects?
I realize the POP 2 is better suited for larger items but the added detail of the MINI can be useful for the study of fine surface details on specimens which the POP can’t pick up.
For the size you are talking about you will generate some large files. On occasion i have file sizes just over 4 GB. I would not want to deal with that with anything under 64 GB of ram, and a good CPU and GPU… I have the Pop2 and a high end structerd light scanner at the moment. Hopefully one of the beta testers will reach out to you soon to help guide you.
You might want to consider using both the POP 2 and the Mini.
The POP 2 has Body mode, which can scan from a distance of up to 40 cm and cover a much larger area at once. Yes, the result has a lower resolution than you might want, but in the photo you attached, there are large areas with little detail in them.
Continue with the Mini to scan smaller areas with the detail you want to capture. Make as many scans as you need.
In Revo Studio, you can use the lower-resolution scan as a frame work upon which to attach the high-resolution scans using the alignment function.
I am official beta tester of MINI , you see you can scan a 0.7mm area with almost 3 millions polygons of data , how bigger your scanned area at once , how less data it will have , so it is recommended to scan maximum 50cm area , and start a new session , and merge them all together . The model will be huge after merging and can strain your computer one way or another , but it depends on many things like for example the pitch point settings or meshing setting .
That is very good, how more features how less issues with tracking , you want a rich features of your object, easy to scan and easy to merge later.
POP2 can pick up very good details , but if you want to go for almost micro details level then MINI will be a better choice of course , but it will take longer processing time for a reason.
as @JeffLindstrom suggested , using POP2 to scan a base model for the cage is a very good idea , as I do that with POP2 scanning bigger objects with body mode that allows you to go beyond 40cm distance to 60cm what is still ok, then after I switch to feature mode to scan finer parts and merge it later together .
So far I did not used MINI yet for scanning larger areas , I made couple of tests but nothing superior yet to show off . It was not enough time yet to test all the possibilities .
@Erked Thanks for your input, I think the large file size will be a necessary trade off for the quality.
@JeffLindstrom That is a fantastic idea, I didn’t know that was possible to combine the scans from both scanners. Your method seems much more practical (and time saving), I agree the items I need to scan do tend to have large areas of relatively less complex geometry.
@PUTV Thank you for your thoughts, the main posts you have made on the POP 2 and MINI are very helpful. I see you mentioned the density of polygons gets lower the larger the object is. Is it correct that a 5 x 5x 5 cm object will have about the same number of polygons as a 50 x 50x 50cm one (as in there is a limit to the max number of polygons, they just get spaced out the larger the object gets thus losing detail)? I watched your MINI android tutorial, have you tried the “fast scan” vs “high accuracy”, would “fast” give a less heavy mesh but still be more detailed than the POP 2?
Do you have a general estimate for the MINI file point cloud file size (in GB)? On average, how many polygons do the high quality meshes generate? I use zBrush for OBJ cleanup.
In regards to the POP 2 for the base and MINI for details, I think this is a great solution which I had not known was possible. As you mentioned with the POP 2, you scan large areas then do additional “feature” scans for detail, do you find it makes the mesh messy with artifacts during alignment? I know many users say not to over scan objects in a rotation, to not go over the same area many times as it can result in a less clean mesh. Do you find that to be a problem when scanning the same parts again in feature mode then meshing with the lower quality base mesh?
I appreciate your feedback, this is just the information I needed.
You’re welcome. I understand, where you are coming from. I hope you can share some of your scans once you get the Mini.
Great question! I would like to know as well.
File size will be determined by the point density and the size of the object you scan. From the sounds of it I don’t think I have to say anything about decimating the mesh, since you use zBrush.
It depends , if you scan fragments and merge them together later you will have as much details as the all merged partials .
This is different when you try to accomplish everything in Revo Scan only , the final mesh will be only 25% at maximum settings , so 75% will be gone . That include the overlapped point clouds and other data that was not necessary.
That why I prefer to fuse just the point cloud and move out to Revo Studio or other programs where I have greater control over everything .
I think I did once , I am not sure at this stage what is exactly the pitch point of point clouds in Android app , I believe it is still the same as before 0.1mm and meshing at 4-5 and the fast scan setting at 0.3mm pitch point , we talk here about a phone , not enough power to process everything , you will need your phone to have minimum 8GB-16GB RAM but there is not way to adjust the settings right now to produce better pitch point , so all just guessing regarding MINI at this moment . The final version will be probably much better , The point cloud created using MINI has much more details since it has higher accuracy and blue light for scanning , but the resolution is the same .
I work only in Zbrush for over 2 decades , I have right now open scan , around 1 million polygons takes around 786MB . I actually never got higher than 8GB from any single scan with POP2 or MINI , unless it was merged.
I can’t give you exact number , as each object is different , it depends of the volume, your settings etc… but I don’t think you will have to worry about . You can handle it for sure.
not really because you are merging point clouds and you removing any overlapped points in the process , I always clean each merged part before merging , usually 1/3 of the data is gone after .
You know in Zbrush we do thing little differently , we collect as much data as possible , then remesh it or create new topology and project the details on a clean new mesh with proper UVs , working with triangles is a pain after all and there. is less functions that can proper handle them .
That is how I told everyone in the beginning to stop overlapping the scans 10 times , making a ball of data that is not easy to handle later and not really necessary. When you merge 2 scans together you only keep the overlaps to minimum and remove all other parts that are not needed , after that you make raw alignment , and after that proper point cloud registration , normals and then mesh .
There are not all functions accessible in Revo Studio so I prefer to do my complex models in Cloud Compare for more precision . I think it gonna be your best friend in your workflow soon after you get MINI and highly recommended .
so where your Museum is located ? I could visit with MINI for a session if around my area.
I see, that makes sense. I plan on using the mobile app to gather the point cloud (if my computer is not available) then do the final meshing on the more powerful computer as suggested.
Ok, good to know. Hopefully as the software is updated and testers have more time we can see some comparisons in this area.
Great, that sounds manageable and not different than my usual work flow. I generally like to gather as high quality data as possible then decimate to make it more manageable later. I use zBrush for all my post mesh cleanup, retopology, decimation etc so I’m used to the larger file sizes.
Good explanation, I will need to actually get the scanner and experiment with the point clouds to familiarize myself with it. I have seen you recommend Cloud Compare in your other post, the results look quite improved in terms of details. I’ll definitely try that out.
I’m not at a museum currently, I usually travel to different institutions to collect data as needed. Thank you for taking the time to answer my questions. Your thorough explanations have been immensely helpful, I’m sure for others as well. It is very appreciated!
