Hello, everyone. Welcome to the webinar. Let's get started. Today's presentation is using
the Dot Product Scanner with Rhinoceros. I am Mary Fugier with Robert McNeil and Associates.
Our presenters are Chris Ahern and Alex Lorman. Chris is the Marketing Manager of Dot Product
LLC. Chris will give us an introduction to the Dot Product Kit. The Dot Product software
turns an Android tablet into a portable real-time 3D capture device. Alex Lorman is the Chief
Technology Officer of C-Machines. Alex is a Rhino and Flamingo expert. C-Machines produces
systems and upgrade kits for autonomous surface vessels. Alex will show us the ease of capturing
as-built 3D data with the Dot Product DPI 8 Kit. The scanned data will then be imported into Rhino
for part design and for analysis after fabrication. And I'm Mary Fugier again with Robert McNeil and
I will be facilitating today's presentation. The presentation will take about an hour to an
hour and a half and then we will open it up for questions. We have a website with the links and
contact information of the presenters. We are ready to begin. Let's go ahead and welcome Chris Ahern.
Hi Chris. Hi and thank you Mary. Happy to be on the line today. I'm going to go ahead and switch over
to my screen here. Again this is Chris Ahern from Dot Product and I'm going to be going through an
introductory presentation on our technology and the Dot Product DPI 8 Kit which is the handheld
3D scanner being used by Alex and C-Machines in conjunction with Rhino. So first off a little bit
about our company and the problem that we decided to solve with the DPI 8 Kit. So we're a global
company based out of Boston, Houston and also Wiesbaden, Germany and we're a privately held
company with our three executive officers shown on the screen here. Raphael is the brains behind
the technology. He really understands how computers understand the world. A computer vision and
augmented reality specialist. He's the developer behind our technology. Brian and Tom on the right
are the ones who saw the application for this technology in the 3D scanning world, engineering
world, design world and more. People who'd be willing to pay for a portable and user friendly
device for instantaneous 3D capture. So the problem that we decided to solve was the challenges of
today's 3D capture solutions. They tend to be very expensive, require a lot of expertise. Miss
occluded areas, you know, you have to go back to fill in more detail or some areas are just too
hard to reach with the large scale expensive solutions. They're time-consuming both in the
field and in the post-processing and not designed for the field worker or the mobile user. So here
comes our solution. The Dot Product DPI8 kit is a complete all-in-one handheld 3D scanner that you
can use to capture 3D from a single-handed device all operating off of an Android tablet. As you
see here, it's a component of three parts. The sensor itself, the tablet and the core of what we
do which is called 5 3D software. It's what pairs these two pieces of hardware and allows for the
real-time accurate 3D data collection. So a quick list of the six things that we're solving with
this device. So first off, it's truly mobile. You can walk around, reach hard to reach areas. Very
simple to use operating again off of a user-friendly Android tablet with very simple software and you
can get up and running in two hours or less and get better and better the more you use it.
Completely handheld, inexpensive at about $5,000 for the total cost of everything needed to start
scanning. Accurate and real-time. So as you'll see in some of Alex's videos, there's a very powerful
real-time feedback that you get not only while you're scanning but you can actually view the data
on the tablet within seconds of capturing your scan. A little bit more on the hardware before
we get into some of the applications. This is the NVIDIA Shield tablet that powers everything from
the processing to the battery life. It's chosen for a reason. It's a very powerful tablet designed
for gaming and it works really well with powering our software and our sensor. The sensor itself is
an RGB depth sensor sending out a pattern of infrared structured light and that is where it
gathers the depth feedback and it also has an RGB camera in the middle that's stitching together
color data onto the depth point cloud in real-time at 30 frames per second. The core of what we do,
as I mentioned earlier, is the FI3D software itself which pairs these two pieces of hardware and
creates an all-in-one solution. What we're getting out of the software is dimensionally accurate 3D,
instantaneous reconstruction as you can see in the video here, viewing your model right after
capture on the tablet, real-time feedback in the green and yellow that you may have just seen and
we'll see more of from Alex so you know exactly how well you're capturing and what you have and
have not captured actually while you're scanning. Real-time feedback and compressed output so it's
an emailable file. We can actually scan and email right from the tablet which anybody who's worked
with point clouds before the traditionally very very large files but our files are usually somewhere
between 10 and 35 megabytes so it's a very small and easy to use both on the capture side and on
the post-processing. And then lastly you have the ability to append data sets together in the field
so you can go fill in gaps, you can capture larger areas by linking data on the fly. So it's
deliverable today with the DPI8 kit, a complete solution that we calibrate and ship out of our
Houston office. All software updates and support are included for 12 months which is important
because as I've mentioned the core of what we develop is the software itself and we're upgrading
it regularly. The detachable mount and protective case are included and the two accessories that it
is compatible with are a mounted light kit so you can capture full color data even in complete
darkness with a diffuse light that is wider in field of view than the camera itself and extension
poles to say capture above a ground etc. And that allows you to actually maintain control of the
tablet in your hands while having the sensor extended at range. Software integration first and
foremost we're integrating directly with Rhino via the free plugin which you can find the link to
in the web page that Mary has already shared. And then here's several other programs who are also
using our our DP file format directly and both the point cloud and CAD worlds. We also export
to industry standard point cloud formats to maximize compatibility with all the different 3D
workflows out there and anything that will play with point clouds to any degree should work with
our product. Some of the professional features that are very powerful with our device as I mentioned
earlier the ability to append so this comes into play in several different scenarios. If you see
that real-time feedback notice you didn't get as much detail on something as you thought you did
or that you missed an occluded area that you hadn't realized during the scan you can actually add
directly back into that same file on the fly in the field. It's a very unique capability and a
powerful application for our users. You can also link multiple scans together on the fly so you
could without any sort of cloud to cloud post processing capture say a whole floor of a house
as opposed to just one room where typically a single room would be one scan with our handheld
scanner you can link four five six seven scans together using that method and capture larger
areas when needed. You can also use targeting for the highest level of accuracy where we
automatically recognized the April tag targets which would be the one on the bottom here which
Alex will demonstrate a little bit during his presentation. Those are completely automatic
improved accuracy just by overlapping over these QR code looking targets during your scan so
there's no manual measurement you don't need any coordinates or anything. It's a very powerful
workload to automatically improve the accuracy of each DPI8 scan with highly accurate loop closure
and automatically recognized targets. Above you'll see a more standard checkerboard target. We also
play very well with these if you want to take a set distance in your scan and specify exactly how
far apart two points are you can reference them that way or you can reference exact known coordinates
say on a job site if you want to improve the accuracy and also reference it to a larger model
for example. All of that is an optional workflow that'll just get you an even higher level accuracy
than you already achieved. Here's a long list of some of the growing applications for our technology.
So everything from construction, BIM and as built documentation to utilities management,
electrical, building and facility management and architecture, real estate, custom design
and installation. So going down this list it's basically any scenario where either you're
already working with 3D and you maybe are or are not experienced with point clouds but also any
scenario where you're replacing what would typically be at best some complex measurements and some
photographs and it may be a sketch pad. With this device you can capture a lot more data in one
visit and just refer back to it whenever you need to pull additional measurements or fit modeling
design into an existing scenario. As I move down the list we also have crime scene and forensics
which is a very obvious application and a powerful tool for police departments today, archaeological
documentation and digital heritage preservation but this is just the beginning so this list
continues to grow as the world of 3D continues to grow. So next I'm just going to go through several
slides of some current applications for our products and some images of the data that's been
captured by our customers today. So this example is from a submarine pier documenting the exact
hookups in three dimensions to ensure that everything fits as needed for the U.S. Navy. Next
you see an offshore facility modification on an oil rig in the North Sea. This represents a lot
of where our technology comes into place. You're designing a new system in your CAD model but you
want to get some quick reliable data of what is actually out there and how it's actually going to
fit into the as-built conditions which are often hard to obtain different than they were on the
original model that you had and sometimes with very limited access to the site which is of course
the case with an offshore oil rig where it needs to be shut down and it's very expensive for every
minute on site. So they were able to capture all the existing space with the DPI8 and then draw
their model directly into the real as-built conditions. Installation verification is very
common as well so similar to the last one however this is a simple bathroom renovation of a dormitory
and they were ordering prefab piping systems in bulk for every bathroom in this I believe it was
a 10-story dorm and they wanted to verify that the systems they were ordering in bulk were going
to actually fit the way they were supposed to before placing the order. So the pepper construction
actually went on site to the prefab facility scanned the prefab system and then brought it back
into their model not to do all of that verification before ordering with this is a pretty unique
application here on a bridge renovation project here in Boston with Skanska and what they're doing
here is using our DPI8 SR kit I should mention there are actually two different variations of
the product and the only difference there is a slightly different version of the sensor so
the standard unit will go from about two to 12 feet in range and the SR kit will let you go from
about one to six feet in range so it doesn't have quite the same versatility but if you know you're
going to be getting you know those more close-up objects or parts exclusively then it allows you
to get a little bit closer get a little bit finer detail on the data so in this project they're
replacing steel base plates on a very old bridge that's being retrofitted and there were thousands
of plates each one with very different unique rivet hole locations for the replacement plates so
instead of having somebody climb up there with a harness tape measure and a sketch pad to jot down
notes they're actually able to use the range pull and keep the men on the ground and capture much
more accurate reliable data where they could then bring it directly into the design environment
and draw their new plates right on top of the point cloud in three dimensions so it's an
application save them a lot of time money and save electrical utility management from underground
vaults to above ground interior systems this has been a very growing market for us and real estate
so especially on the real estate management and renovation side we've got a lot of customers
working in custom installations and real estate development here's an example of a full six bedroom
house captured with our device this is a large-scale project for our device but it's a very powerful
tool and this is all done using the append feature that I mentioned earlier historical
documentation it's very valuable application for preserving three-dimensional color data
of historic site forensic reconstruction as I mentioned earlier so this goes for both accident
reconstruction and interior crime scene reconstruction and then on the bottom left here at the idea of
using point clouds directly with models in order to reconstruct the crime scene and cast molding
so this is a more discreet example of capturing a quick scan with a dpi8 sr kit to get those exact
measurements of a leg and then you send that directly to your design environment where you're
going to be making that mold and lastly just a few more images to show the widespread range of
of where technology is being used from automobile to you know something more toward the consumer
side of things capturing a teddy bear dinosaur and then of course a kitchen renovation project
in the bottom right here so in summary why the dpi8 and 5 3d it's very easy to use quickly deployable
fully real-time tool as I know I've mentioned a few times it's a low-cost solution both on the
equipment itself and the implementation it's very easy to bring directly into your existing
workflows both in the field and in your post processing environment it plugs directly into
your existing workflows and has a very easy upgrade path so that's my rapid fire overview of what we
do and some of our technology and its applications today but most importantly now I'm going to pass
it off to Alex who's going to show us some of his live capture and how he's using the data directly
in rhino today so my name is Alex woman I'm the CTO of steam machines robotics work we're based in
in Cambridge Massachusetts here we we build systems and integration packages for unmanned
surface vehicles so very simply drone boats we use rhino every day from our conceptual rendering
to our design and intent to our as-built and construction drawings for our prototype boats
that we keep here in East Boston here on a cover image that is actually rhino as Flamingo if anyone
interested we spend a lot of time on waves so just a brief bit about steam machines our
autonomous control system ACS has a system architecture that we can deal with direct
operations with a remote control we can deal with other boats a fleet operation style scenario
where you multiple vessels collaborating together some of which are manned some of which are not
it provides for much safer more efficient and more reliable operations at sea for our completely
remote operations we can drive our systems by satellite really any any system that can give
us an IP link we can use our system is more or less caught except for our software obviously
but we use high strength PLC to do our close to our control and actuation so we did have our
naval architect come up with a couple of autonomous hulls in order to show people the future once we
start integrating into into existing boats we can integrate into new boats and allow them to be
optimized for unmanned operations this is a 20 foot by 8 foot so it's containerable very small very
high power work boats that we that was designed in rhino and these these screenshots are out of
Flamingo so if anyone is interested in one of these boats please give me a shout we can
make that happen um but we'll be focusing today on our autonomous control system retrofit onto
our v0 we call it so this in the bottom left you can see it's a 24 foot steel work boat that we
purchased in Holland and had it imported because it's this interesting drive arrangement that you
can see on the bottom right it's a dual asthma stern drive which is a fairly interesting drive
arrangement that's not particularly common in the U.S. that's where you want it to work as you can
see the main image here we've had to retrofit this boat with our control boxes wiring cable looms
cable trays and in mass for our radar and MSP or our lighting systems it's that mass we're
going to talk about today although we have used the dpi-8 in a couple of other scenarios including
hanging upside down in our stern locker imaging the inside of the hull so just briefly a little
bit more about sea machines when we dive straight into rhino we uh our usual concept is collaboration
and oil spill response here where you have an unmanned boat working with a a larger vessel
to tow oil boom across the the surface of the water and it allows workers to stay out of danger
because anyone who's been around crude oil those it's not super fun so they can so workers can
stay safely on the enclosed bridge of the uh the supply vessel and you can use the autonomous boat
that does not require birthing food or or continuous training to stay stay ready you can use that
and just deploy it whenever ready we have we looked at undamned firefighting again this is a
uh a use where you couldn't get human uh drivers in that close and you could potentially stop
a fire before it became a large fire we've not actually built any of these but it is a decent
concept and we've we've looked at remote usb rov collaboration where we would launch an rov from
our from our vehicle and we looked at demo bits from our b0 actually quite shortly and we're going
to work with a company out of maryland to do that again all of these renders are flamingo the water
is uh it's flamingo water that we spent a great deal of time decimating we've also looked at remote
data harvesting a lot of systems subsea have more and more data and more and more subsea
communications in the field and so now we're looking at bringing that back up to the up to top
side so that's a quick overview of sea machines itself as you can see we're going to be looking
at the math so for purposes of expedience we're going to uh show a little bit about scanning
our hull to get an as-built and if you'll all bear with me we have a quick video and i hope you
all can see if the mary will pass around the link to video in case any of you have a a slower
connection or getting any problems with the webinar connection and then you can watch it
either right now later or both so this is scanning the mast in situ on the boat as you can see we've
installed the mast and it has been pivoted back so we could fit the boat into our shop
i will say that one of the limitations of the dpi8 is it doesn't do fantastically well in bright
sunshine and that's a hardware limitation it's nothing to do with the design at that product
so we had to fold this mast down and get it in our shop we had to scan it like this which as you
can see is challenging it can be tricky to see the data so we'll show actually scanning it
with the scanner in real time so again this is this is scanning in real time showing moving
around the the uh the mast and you can see it's kind of a pain to scan because a the access is poor
but b it's a tubular object it's not just flat faces so this is actually me scanning
scanning the mast handheld moving around our boat trying not to put my feet into the bilges
etc and then the uh the video actually cuts out because of premiere pros rendering but if you
want to see the rest of the timelapse it will be posted online and not be a problem so this is
that scan data from that scan after optimization on the tablet so you can see it really does capture
the tubes and it captures the the light mast itself so again a very challenging situation to
actually see good data and you know real mix of materials that mast i was very impressed with the
dpi at that mast is a highly reflective sort of car paint style material and it really captured it
well because it was able to see other objects surrounding it so it didn't just discount the
data uh it's a but it was a reflective reflective thing it did not scan well so we can see this
bit and now just for comparison purposes i just switch on the um the as designed mast so we can
start to see from this point cloud that it really wasn't built anywhere close to what was designed
the pipes are sort of in the right place the angles aren't really correct the side walls here
really just completely wrong and this is why we're going to allow our fabricator to go through
this anonymously and i did a quick scan of of our um of our mast we'll just see if i can yank it up
real quick so anyway you can see that it's it's wrong here and we had to do a little bit of steel
rework on the hinges to get it to fit you can see right here that the hinges don't really go where
they were meant to but now we have gotten it sorted out it works very well we were just out the other
day uh testing the radar and the gps feeds in boston harbour and the mast stayed on and fit well
and when we had to hold on to it due to bashing waves it didn't fall down so all of these are good
things and we uh we really think that the dpi8 has helped us out tremendously we did do a couple
of other works with our stern azimuth gears just scanning the insides of walkers hanging upside
down which was uh certainly interesting now just real quick i'll go through opening a dpi8 file
so let's just open the raw scan of that mast so you literally just drop and drag or hit open
whichever you prefer oh we don't change that that's fine you see here it's going to go through the
frame to open it now the obviously the orientation of this needs a little work that's okay all of
you i'm sure used to doing that right now so before optimization you can see there's there's
doubling there's a couple of artifacts flying around and then after optimization it's really
much much better so this is after in tablet optimization and it returns it to really engineering
and verification level things that uh we can we can use easily so that's that's how we've used the
dpi8 and we again feel like it's it's been an invaluable tool to us especially just on this one
project and i always get to answer any questions i'll hand it back to mary and and chris but i
hope i can answer any questions from a technical rhino side i'm sure chris can many of you have
them from a product side fantastic thanks a lot alex and we can ask anybody to raise your hand
if you have a question or a comment for for alex or chris i have a couple questions in the question
pain and john asks what resolution is the data captured at and as in distance between points
sure i can definitely answer that the the resolution is approximately one to two millimeters between
points and the accuracy you can always tighten it using the workflow that i mentioned earlier
and maintain this across larger distances but it is typically about a five millimeter level
of accuracy and it can actually be even better on the small scale stuff the bridge renovation
project that i showed earlier with the steel base plates and our short range scanner they were able
to get more accurate measurements at a sixteenth of an inch taller than they were with hand
measurements so that's a pretty good representation of the best that can do is five does five millimeters
does that get any better with the use of the april tags or the checkerboard pattern uh yes
absolutely so as i mentioned the resolution is about a millimeter and the accuracy in a
single frame can be as low as two millimeters so using april tags or hard targeting or scale
bar targeting is going to maintain that across frames and across distance and and keep it as tight
as possible so the more that you use the higher accuracy it's going to be and one of the questions
that that i had we'll see if anybody wants to raise your hand and and give us any additional
questions i guess most of these were related to uh accuracy but i was looking for some type of
calibration process um yes absolutely so that's the first question we get from a lot of people
seeing technology for the first time is you know are you using some sort of gps or how are you keeping
the data scaled um and the the way that that's all happening is built directly into our software
and it's all based off of the geometry in the scene and stitching every frame together on the fly
um and basically um it's not using anything outside of the device everything is local
there's no gps no wi-fi nothing else involved but it understands its own look location in relation
to the environment that you're capturing and that's how it keeps everything scaled as i've
mentioned a couple times we have the scale bar workflows where you can take a hand measurement
and say i know that exact distance i want everything to be automatically referenced to that hard
distance and that'll always make things even more reliable and even more accurate um same goes for
the april tags which we recommend just about every time because there is no manual input required
for that other than just putting them in the scene and overlapping them uh either at the
beginning in the end of the scan or or periodically throughout the scan um all those workflows make
it even more reliable and accurate but in general this is the scaled nature of the data uh just
comes from the way that we capture it and stitch it together on the fly so there is no calibration
needed other than what we do to the hardware itself before it ships the specifics um of the
accuracy i just wanted to say it's 1.8 millimeters at a at a one meter distance so that's right
along the lines of the one to two that i mentioned earlier um and that's the the point spacing itself
great great uh details uh chris thank you and alex how long have you been using the
uh dot product scanner in your work i've used it i don't know for about eight to nine months now
something like that for for various uh projects within steam machines that you know retrofitting
a steel boat obviously has a lot of fabrication jobs so we've done more than one with it this
was just the one we chose to show so i've used it consistently for that and a couple of my own
projects and so far it's been absolutely stellar or alex is it hard to to uh go back and uh try
working without it absolutely yeah i uh i would now use it for almost anything over a tape measure
where where you know it's not literally a ruler and i you know it allows me to capture data and
then measure things after the fact i didn't even know i was looking to measure but i've got a scan
of something i can then just look on my computer while i'm designing something rather than having
to go back in the water or go back somewhere else and measure it it's uh it is really an invaluable
tool we do keep it in it in its hard case in our workshop ready for use and do you travel with i
haven't had the need to simply because our fabrication has not moved off site okay but it's
it's uh it's the size of a you know a large hardback book with the uh the pelican box
and sanay's and waterproof so i could very easily throw it in the back of the truck work truck
not a problem at all and take it with okay thank you alex and um garth says uh has a question
chris probably for you are the sensors available or a development that will have higher accuracy
than the two millimeters approximately that you mentioned um yes there are um and uh you know
as we are primarily a software company that's something we're always open to and and under
current development with um and and yeah so we we are working on both you know improved accuracy
but also the daylight capability which is a current limitation as alex mentioned where
you can capture outdoors you know that bridge project is all outdoors but it has to be early
morning late evening in the shade or on a cloudy day so it doesn't work in direct sunlight so that's
another limitation on the hardware side that that we're looking to solve soon as well along
with the the accuracy limitations current i saw a couple other questions from joel you know is it
possible to register with the dpi8 scans with more traditional tripod scanners and it's definitely
yes it's a very common workflow for us we're partnering with a lot of the tripod scanning
companies directly and even with those we haven't we export to those industry standard formats so
we go into uh you know other programs like autodesk recap with a free plugin like mentioned
which that's just a point cloud processing tool so that could be a good gateway into any program
like rhino or others and uh and then we use those standard point cloud formats to work with you know
any other point cloud you get from a different scanner or anything along those lines okay great
and there's a few more questions that are coming in uh is it possible to get a 2d image from the
scan absolutely um it's a it's a matter of slicing through the data that's something that you can do
in various programs i can't speak to um the exact capabilities of rhino as as as well myself but i
know that's a very regular workflow for our customers and you just can you know scale down the 3d
data to exactly what you want um as alex mentioned we have some new software coming out uh as soon
as next week in beta that'll also be operating directly on the tablet or on windows and that
will allow you to perform some cropping of the data before you even leave the field so um it
could be very helpful to crop down to a 2d essentially a 2d scan uh even before you leave
the tablet and then that would be a very very small file and make things really easy um so that's all
coming soon in beta for our customers uh but we're adding more and more to what you can do
so when you bring it into rhino all you have is the the data that you need you don't have any of
those stray points or uh or anything else like that so it makes it really easy um another another
one i noticed here was asking about um annual calibration um we do offer uh an annual or or
biannual calibration for the device to send it back into us but it's really just a matter of how
often you use it and how often you hopefully do not bang it around um but if that is the case
we will recalibrate the device but it's not a set uh amount of time that is required it's just a
matter of uh checking your measurements and if if you're not uh within your specs then you can send
it back for recalibration but it should last for at least a year if not longer um as long as you
don't bang it around too much but it comes in a protective carrying case and is a very compact
solution so it's not a problem did you say uh houston is where the calibration recalibration
would take place okay that's where we do all of our assembly uh production and then calibration
and this might be something that uh a question that alex can jump in at and then we're going to
unmute garth's mic and then he can ask his additional questions about the uh next question
has to do with file size and whether alex finds that the point clouds are unmanageable in uh in
rhino at least for what you're doing alex absolutely not no they and i i see where the
question is coming from traditional point clouds a lot of the laser time of flight scanners you get
gigabyte files and they are absolutely horrible to deal with i've worked with those in the field
before and they're terrible so that scan that i showed of the mast in situ optimized was 16 megabytes
in its native dot dp format so that's really small again that's emailable um once you bring it into
rhino it does expand a little bit it expanded that rhino file about 375 pegs but with rhino 5 without
the uh 4 gig ram limit i've never bumped into a problem i've even you know had multiple scans
together it's it's been no problems at all and vastly easier than working with some of the multi
gigabyte point cloud uh the outputs from other scanners um so how much ram do you have on your
system alex i i personally run a 32 gig of ram system but my system my system's built for other
things any reasonable machine that runs rhino should not have a problem with this okay got it
and it sounds like you do some video editing you mentioned premiere is another product that you run
yeah i i my computer is vastly overbuilt for most of what i do okay nice it's a good situation to be
in okay indeed it is garth garth i'm going to unmute your phone here so we can get your question
about the tag hi thank you i i just wanted to know if the tags needed to be perfectly flat
like on a wall or on a floor if you can uh like wrap it around a pipe something like that uh they
should be perfectly flat um so you can stick it to a pipe but you should um you know even just put
it on a piece of cardboard and stick it to a pipe um it will work uh you know wrap to a pipe but for
the most reliable results we've been told from our developers that it should be a flat surface okay
also i may have been answered but what's the maximum scan volume or like how long can you run a scan
sure um it's it's the best way to describe it is probably in the lens of 20 to 25 feet
by 20 to 25 feet if you're talking in a square area but you can go longer distances for example
but it's usually about the size of a single room uh is is fair description of one scan
and then you can use a pen to link further and further from there using what we call the spider
method where you scan the center of your area of interest and then go out from there so the the
PowerPoint showed a two-story house and that was all captured basically by capturing the staircase
first and then doing multiple of pens back to that staircase to capture each room living
living area etc as an append and how many how many can you append together there is no limit
technically um but the the limitation comes in in wanting to maintain the accuracy of the data
because if you append to an append and append to an append then you know what could be uh you
know a couple millimeters of error can can grow um so basically what i just described there was a
13 scan append where everything was back to that same central scan um and that's pretty much as
far as you would want to go um although you technically could go further it wouldn't be as
reliable thank you hey thanks a lot gars for those questions yeah sure one more thing i wanted to
mention is uh you know as we mentioned in the presentation as well um you know the core investment
is in the software so if we do come out with a more powerful sensor for example or as we already
have in the last year or so come out with a more powerful tablet uh we have affordable upgrade
pass for the hardware it's not like you have to buy an entirely new dpi8 kit every time something
like that happens so it's uh the value is in the software and then uh i also saw a question on that
note about the support and maintenance uh the 12 months that it in the purchase price um so that
is a subscription uh so you get your first 12 months of all upgrade support and maintenance
included and then if you want to stay on upgrades and support after that it is a $700 a year renewal
but obviously whether you renew or not it's it's not a kind of turn off your device it's just to
stay up to date on upgrades which are pretty powerful i was gonna say i see there's one question
here from from sherry about mesh models so just to clarify the the ui of the dpi8 itself
has a meshing engine built in to look at the model more coherently on the tablet you can
actually toggle that between just viewing points and viewing the complete mesh that it's built
it outputs a dp file which is just point so you would need to mesh it and chris do do stop me if
i'm speaking incorrectly but the meshing would be best done in rhino with a meshing plugin if
that's the route you wanted to go i've not personally found that meshing a point cloud is
particularly productive in what i'm doing i find having the points there is much easier to work with
sure and i can speak to that as well um in that uh there are multiple programs that will read you
know not only our dp file format but our other point cloud exports and turn it into a mesh
including rhino with a plugin which i have tested but as alex mentioned the power and point cloud
data you realize is oftentimes not creating a mesh but using it as a reference to your design of
other parts and fittings or is a starting point to build your design around and that's when the
point cloud is very useful even not in a mesh format but it will mesh the same as any other
scan data if that's what you want to do with it and chris what other formats read the dp or what
other programs read the dp file sure so um i'll actually uh chat the link here but um on our website
you can go to software partners and that shows every program that reads the dp format directly
to date i believe it is about um 20 or so um so far and we're trying to get as many programs as
possible to implement it as it is um like i said very small fast and easy to use so here's the link
if anybody wants to click there that will show you the exact and current and if you have the dp
file can you save off a pts file or some other standard format yes that's no problem so it exports
the pts ptx pli ptg e57 vr converter and the question that i think i saw in the thread earlier
might have been referring to our um our new both on and off tablet application dot 3d which is coming
out in beta and that will also in the near future allow you to uh say bring your dp files onto the
desktop or windows based device and then export your pts from there so um a lot of the you know
if you want to maximize your scan time in the field even the optimization which you can do
immediately you can also save that to your back in the office too whether on tablet or off with
our new program so that's all coming very very soon do you have a reseller of dot product in france
we do and our resellers are also listed directly on our website but we'll also be sure to follow
up directly um with felipe for any discussions there and uh one of the questions i think you've
already answered but i thought it was interesting to hear because we didn't get into it in the
powerpoint was the spider technique so yeah and that that's important to mention when using the
append feature um of course the first function for that feature is to fill in data actually into
the same file add more detail fill in an area you missed which is powerful in itself but if you are
using the append feature to link multiple scans together on the fly without using targeting
and capture a larger area um then you basically just think about it quick and you capture the
centermost area of your scan first and then append multiple data sets directly back to that
and that'll allow you to capture much larger areas without stretching out the risk of losing
your accuracy um too much so it was a very easy workflow to capture uh actually a six bedroom
house um with every append going back to that staircase so that's kind of how that works
nice okay great great example and uh as far as i can tell uh i think we have made it through the
question list if you don't see anything else out there alex or chris that you want to comment on
i'll quickly just mention the the very last question here is a coming soon additional partnership
within few points and that goes to the meshing question as well um that's an on-tablet software
coming soon uh from our partners at aliceum and the idea there is that you could export directly
from the tablet some recognized surfaces and pipes um to bring into your modeling environment
uh you know not not just a point cloud for those who may want to to bring those surfaces in so
that's something that's coming soon um and uh it's it's under development uh from the aliceum
okay well maybe we'll do another webinar when that is available so that we can yeah have some
time here with uh our rhino users and dot product users so uh let's um just uh ask alex and chris to
make any final uh comments or closing comments that uh you want to leave our attendees well
i'd just like to thank everybody for taking the time today and please feel free to follow up
directly uh visit our website and then we'll be in touch with some follow-up information
also happy to provide sample data um of different industries uh so you can plug it into your own
workflow and and see how it all works um but thanks for your time and we'll be in touch
i just want to echo that as well thank you all for attending i hope it was at least
moderately interesting and informative and we answered all the questions
anyone wants to be in touch with me directly either about our our robotic boats or about
using rhino in the field uh obviously don't don't hesitate my email is on the on the slides
we'll be posting the full versions of those videos obviously without the black frame the
premier decided to render uh we'll be posting those online and mary will be sending an email
out with those which is very kind of her so thank you all for coming i really appreciate it that's
great thanks uh alex and uh and chris so let's go ahead and move into closing i have the links
also up on the final screen and i want to thank chris and alex for their time preparing this
presentation and for sharing their expertise with this powerful device and on behalf of everyone
here at robert mcneill we want to thank them for their generosity and also thank you for
spending part of your morning with us hope you found the information helpful and interesting
and we'll look forward to seeing you again in a future presentation have a great day
