From 3D Scans to Indoor Navigation and Tracking

yeah i mean uh so reality capture uh or digital twins is way beyond then you know a photorealistic 3d representation of the space uh you can unlock so much value with the existing data that you already have and some of that today we will be showing down the line in this podcast is yeah i mean if you're just let's say creating a scan and just leaving it out there without adding more value you are leaving money on the table And throughout this podcast later today, we'll show you what additional things, use cases you can unlock with the existing data that you have.

I already have had several discussions with different people regarding the digital twins, what is and what is not a digital twin.

And the majority of them said that... simple point cloud or a 3d mesh it's simply another digital twin even if when i suggested that it could be a digital twin of lod zero uh it was not accepted so uh the the difference between just simple point cloud and digital twin should be like like um more obvious uh how would you say like what should reality capturing company do to just move from doing just scans to actually delivering the digital twin.

Yeah.

I mean, there's not many definitions for starters is obviously digital representation of your physical space.

But as you told, right, I mean, it's beyond that, as I mentioned, what else you can unlock? So it could be modeling.

I mean, from the existing digital twin, it could be a live digital twin, which is not just a static 3D representation.

It could be evolving live version of the reality, which gets updated as the reality is changing, right? So I would say it's more if you're just stuck at the 3D point cloud, which is a static, I think that's one step behind.

a digital twin according to me is a live uh representation of any physical space right and and it should unlock more value than just being a 3d representation

uh so is is it right to say that with your solution you can just upsell the already captured data like is it true that you really don't need anything more than just what you have already captured to be to be able to offer something more to your client

Yeah, that is true.

So with multi-set, when we started multi-set, right? I mean, even before starting, when we were thinking of this problem of localization, spatial tracking, we had multiple reality capture, like when we were going and implementing this in venues, they had this, hey, we already have a 3D scan that we did from Matterport like a couple of years ago.

can you use just the existing one and not to do a re-scan because we don't want to do so many re-scans just to enable this particular use case.

And that's where the idea for Multiset was that we wanted to enable, there is billions of square feet of scans already done from varieties of sensor, whether it's LiDAR, but I mean, within LiDAR, there are so many different providers, whether it's 360s, whether now it's Gaussian splats, So with Multiset, the idea is if you already have a scan irrespective of the provider, you can use that scan and unlock spatial tracking, indoor navigation, real-time localization, robotics, and so much more just with your existing scan.

I know that there is a lot of a lot of not fights, but like different diminishing ideas and comments, because the whole.

The whole.

Like the broad variety of techniques, how we can capture the real space, it's going from like 360 cameras up to like high end terrestrial laser scans, the mapping solutions and slimes and everything else.

So there is like a huge difference in terms of quality, in terms of accuracy, in terms of clarity of images and everything else.

So how would you How would you communicate with different clients? Because saying to the guy doing just a 360 video on site with his helmet camera compared to guys doing high end terrestrial scanning for weeks of the same plant.

do you really want to tell them both that guys it doesn't matter to us like both sets could be used for for our solution

yeah so uh i mean as you mentioned right like uh we have so variety of data from a 360 camera to a SLAM-based terrestrial scan or like a traditional terrestrial scan like Metaport and Leicas of the world, right? So obviously the quality of the images, the point cloud, the accuracy matters.

From a multi-site side of perspective, we do give recommendation, hey, if you are doing a small space, you can in fact use a 360 video, you can use a like a low end 3D scan.

But if you're doing like a really large space, we recommend you to go with the higher end of the scanners like Navis, Lycas, even Matterport for that matter.

So we have the recommendation in our documentation.

But what from a multi-site side of things, what we try to do is If you are following that recommendation, the quality of the localization that we produce is very consistent, right? So it's not like, hey, if I have a 360, I'm going to get really bad quality localization tracking.