Magic Leap One is now in the hands of many indie developers. The killer Magic Leap One app can come from one of those indie devs. Indie devs tend to take bigger risks and innovate faster compared to big companies. We see this very often in the gaming industry (Minecraft might ring a bell).
Will MR have the same birth pains as VR?
Even going over the currently available AR app library, I can see that most of the most interesting projects came from small independent developers who came up with very creative ideas. You’ve probably had never heard of those companies before. Yet, these companies or single developers are the ones who helped fuel the AR industry and move it forward. Without them, ARKit would be just a nice name for a promising technology with no actual substance.
The same goes for Magic Leap One. The Magic Leap One will probably goe through the same birth pains as the first virtual reality headsets (e.g. Oculus Rift DK2). No one can ignore this technology’s added value in the professional and entertainment sectors. As for now, it already has proven itself in the business sector, helping to increase manufacturing productivity times fold.
The Lumin OS platform doesn’t limit you to any specific type of application. If you look at the currently bundled applications, there is an audio-visual app (Tonandi), a content creation app (Create), a shooter game (Dr. Grorbort’s Invaders), a social app (Avatar chat, coming this fall), and a home decoration mixed reality app (Helio). I personally think Magic Leap should have released many more ready MR apps, to demonstrate more of this platform’s amazing capabilities, inspire others and increase the conversation about this technology even more,
Developers had a time preparing some ideas for MR
I’m sure many indie devs already have some great ideas listed for this technology and just waited for the Magic Leap One to be released. You don’t actually need to know the exact device specs or the SDK to come up with ideas that can match this technology. I’ve already known more or less what it enables by following some of its demos and Twitch streams that shed some light about what this technology is capable of. Of course, now that the ML1 is released and the runtime environment and SDK documentation are available, developers can start working on some early prototypes and see how everything comes out together.
If I had the Magic Leap One and was developing for it, I would probably spend a lot of time experimenting with the currently available apps just to make that creative transition into that new medium. Magic Leap has mentioned many times that developers need to think differently when developing for this platform, and need to get out of that confined design paradigms that we are all familiar with when we develop apps for standard mobile or desktop screens. This is no longer the case here, the entire world is your canvas now, and that world is different and continuously changing. Your app can be played in many different places, each one is completely different than its peers.
What about outdoor use?
When you develop for the Magic Leap One, you are forced to think outside the box. You have no other option because you are pulled outside your known comfort zone and put into a completely new place where the rules are fundamentally different. Of course, there are many things that overlap, but to make the best out of this platform, you need to take advantage of its unique features, because if not, you just stay in that same spot and bring nothing new and interesting to the audience, an audience who expect new wonders.
After seeing what indie devs have created an augmented reality, there is no doubt in my mind that wonders will happen in Magic Leap as well, and Magic Leap counts on those developers to create something great.
If that process will be similar to VR, it will be slow and it can take a few years until we get to the point where the VR market is right now. However, I believe that AR will see a much more significant growth in the business sector, and this will help accelerate its growth.
I am interested to see how many portions of those newly developed Mixed Reality apps will be for entertainment and how many will be enterprise/business related. There is definitely plenty of open opportunities in the enterprise market, plenty of ideas that have already been discussed and many developers will try to make them a reality. However, with the Magic Leap One (ML1) being aimed for indoor use, that growth potential will be significantly reduced, at least in this current version of the headset.
Take a look at this page on magicleap.care. Let me quote part of it:
“We’ve designed Magic Leap One for indoor use in normal lighting conditions.”
Now, if I am a developer and thinking about making a new mixed reality app for the Magic Leap One, should I just ditch any outdoor mixed reality idea that I add in mind and just focus on apps that can be used indoors? Well, it seems like it after you read those lines. People already tried using the Magic Leap One outdoors. It was actually one of the videos that I’ve seen that I was most excited about, to be honest. I’ve already envisioned the enormous potential of this headset for outdoor use and obviously, I was very disappointed to hear that it wasn’t fully tested and qualified for outdoor use.
What about AR Cloud and the Outdoor App’s Dream?
Magic Leap One has a real-time 3D meshing reconstruction technology that is similar to ones that we’ve seen from some 3rd party AR Cloud providers.I’m still baffled on how Magic Leap is going to integrate this type of technology in and still searching for some answers regarding that. When you read about AR Cloud, your mind is already thinking about great ideas for apps that can be used outdoors, not just indoors.
Let’s take a closer look at the 6D.ai AR Cloud solution and see what this type of technology is so important and later try to understand how Magic Leap One fits in all this.
6D.ai 3D Meshing Overview
This all starts with a real-time 3D mesh reconstruction technology. Essentially, it’s scanning the environment around you and created a 3D mesh of it. This 3D meshing enables several augmented reality features, including occlusion (masking of objects behind physical objects), physics (apply physics to meshed surfaces which represent objects in the real world), real-world interaction (the ability to interact with real-world object via their virtual representation) and procedural augmentation, so you can apply layers of contextual data, which alongside Persistence, that data can be saved to the AR Cloud and retrieved later either on the same device or on another app’s session.
If you were following Magic Leap One’s latest news and coverage on the media, you’ve probably seen that one of the most talked about technologies of that headset is its 3D meshing technology. This technology allows the headset to do essentially what 6d.ai does, mesh the environment and provide a more compelling and believable mixed reality experience. However, 6d.ai was designed to work on regular mobile devices and to allow developers to enhance their current AR apps. As of the time of writing, their solution is currently in beta but fully compatible with the following iOS devices: iPhone 7/7Plus, iPhone 8/8 Plus, iPhone X, iPad Pro 12.9-in (2nd Gen), iPad Pro 10.5-inch, iPad (6th Gen, the one I have).
Here is an AR occlusion meshing was demoed using the bundled Ball Pit Sample Scene.
You can see the real-time meshing working in the background as the user moves the device’s camera around. The speed is pretty impressive if you ask me. We can see that the meshing isn’t perfect and it takes some iterations before you get a reasonably accurate meshing of rounded and complex shapes of objects and surfaces.
A mesh in 6D.ai is built using blocks. A developer can obtain access to that mesh (SixDegreesSDK_GetMeshBlocks), get its size in meters (SixDegreesSDK_GetMeshBlockSize), get the saved-to-cloud status (SixDegreesSDK_GetSaveStatus) and many other different objects and functions that expose advanced functionality that the developer can use in his app.
6D.ai Meshing vs that of Magic Leap One
One of the things that I was particularly interested in knowing is how this 3D meshing technology differed from that of Magic Leap’s, a billion dollar invested company.
The Magic Leap One 3D meshing seems to work fairly fast, similar to what I’ve seen in that Ball Pit demo from 6D.ai. At least judging by a video shared by Avaer Kazmer. I think that 5D.ai is even faster compared to that of the Magic Leap One. Of, course it’s not a fair comparison, because we need to have the same exact scene. lighting, distance from the object and other similar conditions for that speed comparison measurement to be adequate and reliable.
We can clearly see that both the Magic Leap One and the 6D.ai (I don’t know on which device that Ball Pit demo was used with) need several iterations (changing angles over specific objects) to get an accurate 3D meshed representation of some real-world physical objects.
This means that when you run certain apps, it might take time until the scene is reconstructed in a satisfying degree of accuracy needed for a certain app. The good thing about 6D.ai technology is that it can persist the data of the mesh blocks in the cloud and then you can come back later, and that location map data can be loaded from the AR Cloud and relocalized (ensures consistency of the coordinate system across time and devices, thus enabling persistence and multi-user capability) on the same or another device.
In fact, one of the original goals of AR Cloud providers is to create a meshed map of the entire world if possible. Of, course things change all the time, buildings get destroyed, objects are moved, etc, so that p3D mesh representation of the real world should always be updated. When users using different apps, the data from those apps will pa provide a continuous update to the global 6D maps stored in the AR Cloud. This also means greater accuracy for large, small, hidden or complex objects that require many iterations from different angles to be meshed and represented accurately.
From what I’ve heard and seen so far, the Magic Leap One is mostly meant for indoor use, not for outdoor use. However, AR Cloud providers want you to take your device outdoors, they want that precious outdoor landscape to be scanned and they can’t do that by themselves, they need you, users to do seed that data to the AR Cloud for them.
Take a look at this page on 6d.ai. Scroll to the “Brief Briefs” section and take a look at the use cases. You can see: “outdoor large scale”, :Occlusion with cars and people”, “Outdoor epic scale character”, “Outdoor collaborative”, “Outdoor scavenger hunt”, “Outdoor persistent content”, “Outdoor creative/construction”. Yes, outdoor and outdoor everywhere. Why is that? Well, not just because 6D.ai wants to have that largest and most comprehensive 3D meshed map of the outdoor environment, but because most of the incredible benefits of this technology exist for those exploring the outdoor environment or public indoor environments. Some of those use cases include AR outdoors and indoors navigational apps, multiplayer gaming, outdoor art, eCommerce promotional content and dynamic personalized outdoor advertisements, social apps and much more.
Using AR Cloud technologies like that one created by 6D.ai will allow users to use that precise location awareness to anchor data accurately to different objects or locations in the real physical world that provides useful contextual, dynamically changing and even personalized information whenever and wherever you need it.
I’ve asked one person that question that he said that he is pretty sure that the mesh is saved to AR Cloud by default.
That question is still open and I’m searching to get answers to see how this stuff actually works with Magic Leap One. I have no access to the documentation so wasn’t able to take a look at the APIs. If you have that information, please share some information with us in the comment section below. Thank.
I still have so many open questions and the more I read about the ML1, the more questions that I have. I do hope that some of those questions will be answered by Magic Leap One developers or Magic Leap officials. I will update my twitter for any insight or answer that I receive, so all of us can get a clearer picture of how this technology works, it’s advantages and disadvantages.