If today I were considering buying the Revopoint scanner (as it appears in the intro) and saw this video, I would most likely reconsider purchasing that scanner... I suggest removing the Revopoint scanner from the intro because it can't be compared at all to what you presented using your scanner! I've been using the Revopoint POP2 for over a year. The scanning software is continuously updated and improved. There are almost no tracking issues with it! The stitching of scan fragments also works very well. With default settings and a black tabletop, you can scan the entire object by rotating it because in this setting, the scanner doesn't detect black. One last thing - textures. This is the biggest flaw of this scanner, although it's not as bad as it might seem. They may not be suitable (not always!) for production needs (don't compare it to photogrammetry!), but for the second or third plan - absolutely.
I’ve inspected several POP2 scans and tracking issues aside, I don’t think they’re any different in quality than the Seal. Successful scans from both scanners will look very similar. It was one of the reasons I didn’t go for the pop 2 or even the pop 3 and decided to give the new product a try. Let’s see though what the future holds. I’m cautiously optimistic.
So from watching a lot of your information is wrong, complex objects align easier due to complex geometry, turntable mode works great assuming you are scanning something with asymmetrical features but there are work arounds for this like tracking markers and or random objects in the frame to keep tracking like nuts bolts hot wheels or specifically designed tracking bases. This is not the best scanner in the affordable range. There is a learning curve to doing this well, making a video about how easy or hard something is based on limited experience but the same can be said about having too much as to that person its easy.
Great little review. Sadly a lot of these scanners get a hyped marketing campaign promising the world only to find out that the claims are overstated by a long shot. I feel these companies are tricking people into backing these products on Kickstarter and are then shortly abandoning it for another Kickstarter of their new scanner promising the same things they already promised last time. Great way to get money to do R&D with the money of actual end users. I find this behaviour rather shameful and feel bad for all the people falling for these ...
Agreed 100%. The crowdfunding campaigns of all these players is definitely a weird approach. And most of the times the target is a very low amount. 20K-60K somewhere around that area.
Dimitris, I really enjoyed the video and found it very useful. It’s been around 15 years since I was regularly 3D scanning objects and doing photogrammetry with what now look like Stone Age tools! 😄 Future Technology Question: Are you aware of any AI and/or ML assisted tools that are available or being worked on that would allow for taking a group of many different photos of the same type of object and constructing a 3D model? My specific use case would be taking a large collection of bird photos of one particular species and having a system be able to extract the bird from each scene and use the various angles and poses to create a facsimile of that bird? With animals, especially wild ones you’re not going to get them to pose for a scan or well structured photogrammetry captures. I wouldn’t expect extremely high fidelity by the way! Anyhow, just a really wild wish that may not be practical any time soon, but I thought I would ask. Keep up the great videos! 👍
Oh boy! photogrammetry 15 years ago? That must have been torture! 😆 About your question: Even though there are AI tools currently that can take a photograph and produce a 3d model, there's nothing yet that does what you want. I wouldn't be surprised though if something like that comes out in a year's time! Things are moving ridiculously fast in the A.I. world!
that pinecone will work with manual alignment but you have to place around auxillary objects at least 4-5 then the aux objects will help the alignment the main issue the precision and resolution is solved beyond that artifical intelligence never going to be comparable with human so overall this scanner is a true deal just a remark that the nextengine scanner wasnt better however with all sofware and turn table costed $ 5500 likely 12 times more than this scanner and that has also much issues with alignments let alone the artec scanners costing 30-50 times more for not a lot better result
You should have get the EinScan which also plays in the low budget segment but actually deliver on the software due to the legacy. I don't see a point of compareing a proper Photogrammetry setup with a none proper 3d scanning setup. That's unfair.
You can compare the two because for the same amount of money spent on a budget scanner you can get a nice photogrammetry setup. The seal scanner is around 700 euro, For that amount of money you can get a used camera and a turntable that will give you infinitely better results. I would have loved to try the einscan scanner but for some reason they don’t have Mac support. I know they’re planning to add it but for now it’s PC only.
@@MrGTAmodsgerman Even without the lights and polarizing filters you will be able to get great results. Of course it’s better to have those too but even without them you will get amazing results compared to a budget 3D scanner
That's why Elon Musk said lidar is useless and simple "vision" is way more effective. Lidar is good for multipass in survey where lasers pass through foliage while camera can't see them IMO.
I just got the same scanner as a early backer and have it now here since 3 weeks. I have to say the results are pretty disappointing if you compare it to their promotional videos. Non of the scans went flawless and I never got a proper scan result. Some areas looked promising, but no model was finally good enough to use. I didn’t expect high end results, but wanted something, which could generate some quick results for the backgrounds. But in my opinion, every scanner / photogrammetry app worked better for me. The meshes sometimes weren’t so precise, but good enough to put the product in the background. But most of all: it was just finished with one attempt. With the seal, I tried several scans and setups but never got a problem finished scan out of it. Also the limited size of the product/area you could scan is a negative aspect. The range of products, fitting is very limited. I would definitely not recommend buying this scanner or anything close to it.
If you’re willing to sink hours trying out stuff with the Seal, you will be able to improve your scanning success but for sure it can’t beat even the laziest suboptimal photogrammetry session in the world :D Matter and form and Revopoint already have their new scanners on kickstarter, so on the positive side things are progressing. Given the current development pace, I’m willing to bet that in a year’s time we might have a budget scanner in the 1K to 1,5K region, that manages to produce ok-ish to good results. But let’s see. Until that time comes it’s photogrammetry all the way for me.
If today I were considering buying the Revopoint scanner (as it appears in the intro) and saw this video, I would most likely reconsider purchasing that scanner... I suggest removing the Revopoint scanner from the intro because it can't be compared at all to what you presented using your scanner!
I've been using the Revopoint POP2 for over a year. The scanning software is continuously updated and improved. There are almost no tracking issues with it! The stitching of scan fragments also works very well. With default settings and a black tabletop, you can scan the entire object by rotating it because in this setting, the scanner doesn't detect black.
One last thing - textures. This is the biggest flaw of this scanner, although it's not as bad as it might seem. They may not be suitable (not always!) for production needs (don't compare it to photogrammetry!), but for the second or third plan - absolutely.
I’ve inspected several POP2 scans and tracking issues aside, I don’t think they’re any different in quality than the Seal. Successful scans from both scanners will look very similar.
It was one of the reasons I didn’t go for the pop 2 or even the pop 3 and decided to give the new product a try.
Let’s see though what the future holds. I’m cautiously optimistic.
So from watching a lot of your information is wrong, complex objects align easier due to complex geometry, turntable mode works great assuming you are scanning something with asymmetrical features but there are work arounds for this like tracking markers and or random objects in the frame to keep tracking like nuts bolts hot wheels or specifically designed tracking bases. This is not the best scanner in the affordable range. There is a learning curve to doing this well, making a video about how easy or hard something is based on limited experience but the same can be said about having too much as to that person its easy.
Thanks for the review. @10:00 When you say it looks better in PG (photogrammetry), which PG software did this, and how many photos did you take?
Hey there! I used Photocatch. It’s a small little software for the Mac, that uses Apple’s photogrammetry SDK. I used 225 photos in total.
Great little review. Sadly a lot of these scanners get a hyped marketing campaign promising the world only to find out that the claims are overstated by a long shot. I feel these companies are tricking people into backing these products on Kickstarter and are then shortly abandoning it for another Kickstarter of their new scanner promising the same things they already promised last time. Great way to get money to do R&D with the money of actual end users. I find this behaviour rather shameful and feel bad for all the people falling for these ...
Agreed 100%. The crowdfunding campaigns of all these players is definitely a weird approach. And most of the times the target is a very low amount. 20K-60K somewhere around that area.
Dimitris, I really enjoyed the video and found it very useful. It’s been around 15 years since I was regularly 3D scanning objects and doing photogrammetry with what now look like Stone Age tools! 😄
Future Technology Question: Are you aware of any AI and/or ML assisted tools that are available or being worked on that would allow for taking a group of many different photos of the same type of object and constructing a 3D model? My specific use case would be taking a large collection of bird photos of one particular species and having a system be able to extract the bird from each scene and use the various angles and poses to create a facsimile of that bird? With animals, especially wild ones you’re not going to get them to pose for a scan or well structured photogrammetry captures.
I wouldn’t expect extremely high fidelity by the way! Anyhow, just a really wild wish that may not be practical any time soon, but I thought I would ask. Keep up the great videos! 👍
Oh boy! photogrammetry 15 years ago? That must have been torture! 😆
About your question: Even though there are AI tools currently that can take a photograph and produce a 3d model, there's nothing yet that does what you want. I wouldn't be surprised though if something like that comes out in a year's time! Things are moving ridiculously fast in the A.I. world!
that pinecone will work with manual alignment
but you have to place around auxillary objects at least 4-5
then the aux objects will help the alignment
the main issue the precision and resolution is solved
beyond that artifical intelligence never going to be comparable with human
so overall this scanner is a true deal
just a remark that the nextengine scanner wasnt better however with all sofware and turn table costed $ 5500 likely 12 times more than this scanner
and that has also much issues with alignments
let alone the artec scanners costing 30-50 times more for not a lot better result
You should have get the EinScan which also plays in the low budget segment but actually deliver on the software due to the legacy. I don't see a point of compareing a proper Photogrammetry setup with a none proper 3d scanning setup. That's unfair.
You can compare the two because for the same amount of money spent on a budget scanner you can get a nice photogrammetry setup. The seal scanner is around 700 euro, For that amount of money you can get a used camera and a turntable that will give you infinitely better results.
I would have loved to try the einscan scanner but for some reason they don’t have Mac support. I know they’re planning to add it but for now it’s PC only.
@@marvelousdecay For 700€? But what about polarisation filters and lights and such. Those are making a proper photogrammetry setup 🤔
@@MrGTAmodsgerman Even without the lights and polarizing filters you will be able to get great results. Of course it’s better to have those too but even without them you will get amazing results compared to a budget 3D scanner
@@marvelousdecay Ok, thanks for the explaination. Have a great Christmas!
@@MrGTAmodsgerman You too!
That's why Elon Musk said lidar is useless and simple "vision" is way more effective. Lidar is good for multipass in survey where lasers pass through foliage while camera can't see them IMO.
I just got the same scanner as a early backer and have it now here since 3 weeks. I have to say the results are pretty disappointing if you compare it to their promotional videos.
Non of the scans went flawless and I never got a proper scan result. Some areas looked promising, but no model was finally good enough to use. I didn’t expect high end results, but wanted something, which could generate some quick results for the backgrounds. But in my opinion, every scanner / photogrammetry app worked better for me. The meshes sometimes weren’t so precise, but good enough to put the product in the background. But most of all: it was just finished with one attempt. With the seal, I tried several scans and setups but never got a problem finished scan out of it. Also the limited size of the product/area you could scan is a negative aspect. The range of products, fitting is very limited.
I would definitely not recommend buying this scanner or anything close to it.
If you’re willing to sink hours trying out stuff with the Seal, you will be able to improve your scanning success but for sure it can’t beat even the laziest suboptimal photogrammetry session in the world :D
Matter and form and Revopoint already have their new scanners on kickstarter, so on the positive side things are progressing. Given the current development pace, I’m willing to bet that in a year’s time we might have a budget scanner in the 1K to 1,5K region, that manages to produce ok-ish to good results. But let’s see.
Until that time comes it’s photogrammetry all the way for me.
Don't buy. Summary.