Magic Leap's Augmented Reality Tech On Display?

What do you think? Is this real or fake? This is supposedly a "game we are playing around the office right now" but elements of the video (in game weapons casting real shadows on the player's hand, fake recoil, Weta involvement) look very fake.

Unfortunately, we couldn’t make it to TED, but we wanted to share one of the things that we’d planned to share at the talk. This is a game we’re playing around the office right now (no robots were harmed in the making of this video).

Click to expand...

As real as Dolly Parton's boobs.

I can see a great market for 2D augmented reality in aviation though, as it would be a great way to three dimensional "skyway" lanes for aircraft to fly in, as well as showing the past and projected flight paths of other aircraft in your field of view using a HUD or glasses.

Clearly fake. The Weta Righteous Bison replica isn't going to make your arm move like that.

Maybe . . . just maybe, seeing the gun do that is what makes it fire, but I doubt it.

Odds are there's some version of this in reality, but this is clearly a highly edited promotional video.

makes you realize why they don't like Hololens. they are a direct competitor,lol

Would be cool to have a solenoid in each of the guns. When you pull the trigger it kicks back. That would add a really nice tactile element.

Whether that is fake or not (it's fake), what is this fascination with augmented reality? There's so much to see in front of your eyes, why blind it?

Broad approximations of hand position are all it takes for a demo like this, unfortunately, we are going to need finger and hand tracking while in vr, and leap can't do this within an acceptable threshold.

So yes, they are showing something they can do as best as they can, and it's essentially the upper limit.
Unfortunately vr needs more.

the-one1 said:

Whether that is fake or not (it's fake), what is this fascination with augmented reality? There's so much to see in front of your eyes, why blind it?

Click to expand...

Not sure if you're serious, or if that's a deep philosophical question. However, you're on a site that's dedicated to new technology, high-end home computing and GAME PLAYING. This is a new way to play games, which is exciting all on its own. People play things like laser tag, paintball, airsoft, etc. Imagine having a little prop gun, a set of glasses, and then a whole park or forest to play in. The glasses map new types of environments and objects into what's there.

Or like the Microsoft version, playing Minecraft in your living room. Like Lego, but you have an infinite supply, can save your creations, have others step into your creations, and no cleanup. Sounds like fun to me.

I can even see some very cool engineering applications for this. Walk-throughs of facilities that are being built, springs to mind.

There are quite a few possibilities here. It doesn't have to be black and white. It doesn't have to be a replacement for reality. However, if you're going to play a game, why not play it this way if the game suits it. An FPS where you can actually move around freely, or stare at a monitor? Even that doesn't have to be mutually exclusive though. Some things will lend themselves to the tech, while others are still better in traditional terms.

Looks real enough. I'd buy it for a dollar. I'd rather shoot real robots than play some VR expensive bs.

I'll just wait a few more years until the the 'bots make us into some kind of towering human-battery system, which was proven illogical by some terrible movie from the 90s, so that I can then break out of and then go on a wild bullet storm. Naturally, I will have to have an elite team of whatevers to help. I'll make sure to recall the glory days of VR during my hot lead festival. /sigh, (diety) I am bored.

I'd rather have an elite team of whatsits rather than whatevers, but.... whatever...

J3RK said:

People play things like laser tag, paintball, airsoft, etc. Imagine having a little prop gun, a set of glasses, and then a whole park or forest to play in. The glasses map new types of environments and objects into what's there.

Click to expand...

That would be nice, but definitely stick to a indoor lasertag type environment. Otherwise, where do you house the network connectivity and gaming computer and everything while running around a vast area, you'd look like a retard to anyone watching, and if wearing headphones and using a prop gun I can already see you getting shot by cops, lol!

Would also be nice in the bedroom, if you could scroll through a selection of victoria's secret models to superimpose over your frumpy "big is beautiful" girlfriend, lol! That's what you call enhanced reality, taking the paper bag to the modern age!

More then likely the hand phaser has some sort of blue tooth or wifi device inside it that includes position information and trigger pull information.

Ducman69 said:

That would be nice, but definitely stick to a indoor lasertag type environment. Otherwise, where do you house the network connectivity and gaming computer and everything while running around a vast area, you'd look like a retard to anyone watching, and if wearing headphones and using a prop gun I can already see you getting shot by cops, lol!

Click to expand...

I had more of a group setting in mind. If a whole group of people look like stupid, it's less stupid. Or something like that anyway. Also, yeah, maybe not a vast area, but maybe a walled in forest area like paintball would still be good.

Its def. fake but lets take it as a sign of things to come.

With a "kinetic" 3d spatial scan camera, a leap motion sensor and a HUD you could def. see something like this becoming possible in the future. Even things like shadow cast from virtual objects although properly integrating real and virtual objects (like hands gribbing items) are extremely tricky.

And ofcourse without knowing the layout there is no way something can be behind something else.

Actually looking over the video again I realized that it might not be fake, we however misread it.

The video isnt about a injecting virtual objects in the real world but actually about injecting real items (in this case the players hands) in the virtual world.

If you look at the video closely you def. can see that the environment is rendered and the hands "created" by scanning the persons hands, this also explains how the hands can interact with virtual objects (they are actually 3D hands/arms that represent the persons hands).

They hide the "digital" feel of the hands by blurring them a lot in the video tho.

Not even close to what they are working with...

Maybe you should say "supposedly" working with and yes I have read those articles as well but there are discrepances in the video that either makes it fake or more inline with what I suggested.

Gizmodo did a write up on the Magic Leap company. Pretty neat stuff.

Looks like they are using team fortress 2 as the base.

New model for a turret.

Dr. Gorb's blaster.

Robots out of mann v mann.

I don't see why this wouldn't be possible with today's technology. occulus rift, leap motion sensor, or re-purposed kinect. kids at MIT do this stuff all day.

GoodBoy said:

Looks like they are using team fortress 2 as the base.

New model for a turret.

Dr. Gorb's blaster.

Robots out of mann v mann.

I don't see why this wouldn't be possible with today's technology. occulus rift, leap motion sensor, or re-purposed kinect. kids at MIT do this stuff all day.

Click to expand...

Its tricky since a HUD is a display on top of the real world, placing an item behind something like a table which you can see underneath isn't really possible since you always render in front of objects.

Its like a picture on top of a picture, intermingling the two is really hard. Increasingly so if you want things like shadows to be rendered accurately.

Even if you do a scanner for the 3d space and get a rough 3d environment mapping itll never be accurate since you cant see behind things (see kinect 3d space modeling such as https://www.youtube.com/watch?v=Nsljftc4Z1I for a reference to what you typically can get).

Thus its a lot smarter to start with a 3d model and impose real object instead of the other way around, you can pair knowledge of full 3d environment with simpler less functionality in scanned 3d spaces.

This ties in well with the expertise the Magic Leap people have, scanning, image recognition, 3d modelling etc.