Magic Leap raises $827M in Series C • Hacker News Board

Re: Magic Leap raises $827M in Series C

#171

Post by woodman » Thu, Dec 10, 2015, 5:42 PM UTC

ansgri wrote: post #160

woodman wrote: post #153

Earlier quoted context omitted.

I'd be very surprised if a modern cpu couldn't handle the task, especially if you were clever about detecting regions of interest, predicting head movement and cache maintenance. But I'd also be surprised if they go to market with an x86 under the hood. I remember reading a while ago about how smart tvs were using ANNs for upscaling, so it has been done at scale. rimshot

(1) TVs don't have strict latency requirement. I've hard latencies of 100 ms are common. (2) Upscaling ANNs process rather small image neighborhood radius, and required processing power is on the order of O(r² * log r), and if a minimally recognizable cat is 50x50 px and for upscale you use a very large window of 16x16, that's 14 times already.

Latencies of 100 ms may be common because TVs don't have strict latency requirements.

16x16 is a very small window, I have no idea what they're using for TVs, but 128 isn't uncommon in post production ANN upscaling. Also consider the fact that ANNs have not received anywhere close to the level of attention in optimization that compilers have, so there is also a lot of potential slack to be taken up if real time processing demands it.

Re: Magic Leap raises $827M in Series C

#172

Post by rasz_pl » Thu, Dec 10, 2015, 5:43 PM UTC

x0x0 wrote: post #79

rsp1984 wrote: post #70

I've been working on AR and related technologies for almost the last decade and I've been part of the first handful of people working on Google Glass. Bottom line I've seen a lot of promising AR technologies come and go. My personal take on this is that they may indeed have some very good, if not revolutionary, display technology. However : The big, big obstacle to delivering credible AR is latency. Contrary to VR, t…

Since you're an expert: what about these videos is hard? The things that jumped out at me are: 1 - the robot moving behind the table leg (ie you have to do depth recognition of objects in the scene) 2 - the user's hand interacting with the artificial elements in the scene. Some code had to recognize a hand and figure out which element it was touching. What strikes you as the hard parts of those videos besides the rea…

1 or have premade 3d environment model and do accurate position tracking. Position tracking is a LOT easier to do realtime.

2 bullshit CGI "this is how we hope it would look like if it was real" demo

Few months ago their apparatus was one color only, stationary and the size of a desk. Now all of a sudden can be strapped to a camera and does colors? color me sceptical :(

Re: Magic Leap raises $827M in Series C

#173

Post by jlarocco » Thu, Dec 10, 2015, 5:57 PM UTC

jodrellblank wrote: post #146

jlarocco wrote: post #85

Earlier quoted context omitted.

There's a big difference between being really cool and being something people will pay for. The tech demos are undeniably, but after "Wow!" my second thought was, "If I'm not a gamer, why would I buy this?"

Because they will be able to put ads in front of your face all the time. And that means that first, they'll make it do enough desirable things that you will accept ads. And those will make you want it.

That's not a very convincing argument and doesn't answer the question at all. I already can't think of a reason I'd buy one, and telling me it'll be full of spam isn't helping.

I'm also a little disappointed they've raised so much money if their long term goal is just spamming people in "augmented reality".

Re: Magic Leap raises $827M in Series C

#174

Post by mondoshawan » Thu, Dec 10, 2015, 6:04 PM UTC

rsp1984 wrote: post #169

mondoshawan wrote: post #162

Earlier quoted context omitted.

I, too, was part of the first few folks to work on glass. Now I work at ML. Name here is June. We should talk. :-)

Cool, you have an email or Skype name?

june@theonelab.com -- feel free to chat me up on Hangouts.

Re: Magic Leap raises $827M in Series C

#175

Post by aetherson » Thu, Dec 10, 2015, 6:19 PM UTC

pravda wrote: post #128

paulftw wrote: post #80

Earlier quoted context omitted.

> Perhaps they get around that valuation with 100M/year in patent licensing. Patents expire after 25 years, so at 100M p.a. deal for investors is spend 1B to get $2.5B back, two decades later. I'm not sure that even beats fixed interest. Even with continuation it's still around 10% return. As others have pointed out O(1B) is same as O(1).

>Patents expire after 25 years, Twenty years. At least in the USA. Carry on.

And they start from time-of-approval (or filing?), not when you first put together a salable proof-of-concept, so for patents like which are not trivially possible to implement, you won't get all 20 years of it.

Re: Magic Leap raises $827M in Series C

#176

Post by shpx » Thu, Dec 10, 2015, 7:05 PM UTC

shostack wrote: post #97

geori wrote: post #49

If this turns out to be the successor to smartphones, then these comments will look like people laughing at the Instagram acquisition.

Funny you mention that. Was just thinking about the book "Rainbow's End." In the near future, it had most people interacting with tactile "muscle-movement" based input, and contact lens displays. Wearables/wetwear are almost certainly going to be the future at some point. Our grandchildren will laugh when we tell them about having to hold a device in your hand with a tiny screen where you hunt and peck at an even tin…

I might sound like I'm joking, but as soon as I get an AR headset I'm building a computer vision system to detect billboards and overlaying a black rectangle over top. It's already possible to fill in items of a given class in images with deep learning, and it'll only get better.

Re: Magic Leap raises $827M in Series C

#177

Post by RightWingRabble » Thu, Dec 10, 2015, 7:27 PM UTC

sixQuarks wrote: post #130

michaelbuckbee wrote: post #42

The money flowing into Augmented Reality in general isn't so much about 'augmented reality' as a thing today so much as it's a big audacious bet on it being the next wave of technology that replaces mobile phones.

I believe smartphones will be looked back from the future and seen as what they really are: a clunky way to interact with information. AR is definitely the future of all interfaces, not an audacious bet at all, pragmatic in fact.

Meh, I'm not so convinced. AR is certainly a great way to interact with real time information, but there's a time and a place for that. I don't want to be out at dinner and be interrupted with notifications while I'm having a conversation. Laptops and Mobile phones have one huge advantage that AR and even desktops don't have: they're easy to put away. Close the laptop, put the phone in your pocket or in a drawer and it's gone. I suppose you could do the same with glasses, but they're still more intrusive.

Re: Magic Leap raises $827M in Series C

#178

Post by icedata » Thu, Dec 10, 2015, 8:31 PM UTC

x0x0 wrote: post #72

sigmar wrote: post #34

Earlier quoted context omitted.

>To be raising such ridiculous amounts, they must have some incredible technology to demo. I used to think along this line too. But examples like Theranos make me more skeptical

Theranos has raised a little less than $90m according to crunchbase. These folks have already raised $592m.

According to Fortune they have raised around $750M http://fortune.com/2015/10/28/theranos-is-seeking-to-raise-2...

Re: Magic Leap raises $827M in Series C

#179

Post by toby » Thu, Dec 10, 2015, 9:37 PM UTC

oskarth wrote: post #38

toby wrote: post #32

Earlier quoted context omitted.

That would probably get shut down pretty quickly, as it would be essentially allowing a group of small-time players to make non-sports-related proposition bets.

What would be wrong with that? People do it informally all the time.

Wrong? Nothing. It just tends to attract the attention of regulators when done at scale. Even Nevada doesn't allow casinos to run proposition bets on anything but sports.

Re: Magic Leap raises $827M in Series C

#180

Post by shostack » Thu, Dec 10, 2015, 10:15 PM UTC

shpx wrote: post #176

shostack wrote: post #97

Earlier quoted context omitted.

Funny you mention that. Was just thinking about the book "Rainbow's End." In the near future, it had most people interacting with tactile "muscle-movement" based input, and contact lens displays. Wearables/wetwear are almost certainly going to be the future at some point. Our grandchildren will laugh when we tell them about having to hold a device in your hand with a tiny screen where you hunt and peck at an even tin…

I might sound like I'm joking, but as soon as I get an AR headset I'm building a computer vision system to detect billboards and overlaying a black rectangle over top. It's already possible to fill in items of a given class in images with deep learning, and it'll only get better.

Agreed. I personally can't wait for the inevitable ad blocking war that ensues.

Unfortunately, if I had to posit a guess, it would be that the main "application space" for AR won't be the free and open web. It will be the walled garden appstore(s). You'll have your staple apps, but they will all be self-contained, and the owners of the hardware/OS will have strict controls around the ability to do things like, block ads from their ad network (which might be the only one allowed on the device).

Big tech companies don't like the users being able to control what they see through a browser that can install whatever plugins the user wants. Make no mistake, they want full control over the experience, and have no qualms about doing whatever it takes to own that experience. To not do so enables their competitors to claim a foothold in their territory.