On the other hand. Because of the low price you can also just wait and see if/what they deliver and get It at the retail price if it's any good.... atm you missed the super early bird anyhow....
the worst thing about Kickstarter for me is the delivery time, I am almost sure that between now and June next year there will probably be a better scanner available
So true Xeno
So true Xeno
I was planning on buying a Einscan-S scanner, but the price was a lot more, and was too late to get the early bird special, but I heard they were already working on a more accurate version,
and will probably be another Kickstarter 4+month project, until then I can experiment with the Eora scanner.
and maybe the Eora scanner will get even better with the large amount of pledges.
I don't believe their precision. It's very, very, very unlikely that they will get below the 100 micron precision. I've made a laser scanner with fairly expensive (industrial grade) computer vision cameras and I wasn't able to even get -near- the 100 micron precision. Let alone what they claim to get.
If you look at the 3D models you can clearly see that they don't reach the 100 micron precision. More like 300-400 micron.
My first idea for a 3D scanner was actually this, but I scrapped this idea for a number of reasons;
1. If you move your phone (pressing the interface, getting a call, whatever) you need to recalibrate. Laser scanners can take quite a time to calibrate.
2. If you want better scans, you need a phone with better camera / lens. Buying a whole new phone just because you want a better scan?
3. Support -hell- Everyone has a different phone, with different specifications (so they will get different results / issues/ etc). Also note this in buying; Your quality will greatly depend on the quality of your phone!
4. Laser scanners are known for being slow. Very slow. Both because they need a -lot- of images for a proper scan. Their setup makes it even worse because they use a phone. Edge detection (if you don't want a completely naive solution (Something more complicated then thresholding like Canny or Hough)) is a nightmare. Using the naive approach will work, but only in lab-like conditions (and on certain objects).
They use a moving laser, which means that for a single snapshot, you will need about 400-500 images. Due to syncing, you can capture about 4-5 frames a second. For a full scan you generaly want about 12 snap shots. This means a full scan will be about 20 (!!) minutes (compared to a projector based setup, that can do the same in 2 minutes).
They are also using a color camera, which will impact the quality of the model due to the Bayer filter in the camera. This is also why they use a green laser; Color cameras are twice as good with green light as with other colors.
I do have some good things to say about this though;
It's cheap for what you get. It's also a step up from regular laser scanners, as the laser can rotate (which increases the number of objects that you can scan).
Recommended Posts
xeno 108
I backed it, I was already looking into 3D scanning, and this was the best resolution for a small price.
and because of the low pricing ( for a comparable scanner), it won't be a huge problem, if they won't deliver.
backing a kickstarter is always a risk,
the worst thing about Kickstarter for me is the delivery time, I am almost sure that between now and June next year there will probably be a better scanner available
Link to post
Share on other sites