Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Oculus Rift hack transfers your facial expressions onto your virtual avatar (arstechnica.co.uk)
91 points by 51Cards on May 21, 2015 | hide | past | favorite | 36 comments


This is a fundamental technology that will be core to social VR. I really hope they nail it.


Hi, i m the lead of this research project which was really more of a first prototype and feasibility test. We have already started working on the next version though, but can't comment on it yet.


Even something as simple as head gestures, which can be done with just about every currently-known HMD, is hugely helpful in expressing emotion. Check out AltspaceVR, it's pretty incredible. I had an hour-long conversation with someone and even after the first 5 minutes it felt like we were just sitting in Halloween costumes, talking to each other.


I have tried Altspace many times :) We've built something similar in some regards. We are helping people learn foreign languages in immersive environments (including VR): http://learnimmersive.com


It would be really interesting to see something like this mixed with commercial software like FaceRig (http://store.steampowered.com/app/274920/).


There are a lot of players who are trying to commercialize facial expression tracking. Notably, http://affectiva.com https://www.youtube.com/watch?v=AbO0Po8Mgis

disclaimer: I work for them.


Have a look at https://www.kickstarter.com/projects/fove/fove-the-worlds-fi...

They are working on a similar technology.


Eye tracking looks very promising. After seeing the video, I'm curious to see how they plan on adding facial expression tracking.


I think face tracking is planned for the version after the Kickstarter.


They will probably add some IR or optical camera beneath the headset. Not quite sure if they can also pick up some face expression by using the eye-camera as well (like smiling).


One step closer to the Metaverse.


If I were William Gibson, or even Neal Stephenson, I'd be sitting back right now saying, "yup. Told ya".


Except they've been wrong almost universally in their career from a predicative POV (which is a bad way to judge sci-fi, btw. They write dramatic narratives, theyre not futurists trying to predict future tech, they're storytellers). The reality is we don't know if this stuff will ever catch on. Mounting a big black box on your face is limiting and unappealing to most. Hardcore gamers? Probably will deal with it. Everyone else? Probably not.

There's also a fundamental issue here that's rarely addressed: that putting meatspace conventions into virtual spaces rarely works out. I don't want to have to wake up, dress, shower, etc my avatar so she can get in a car and drive to the virtual mall to walk into the virtual store to buy stuff. I just want to load up amazon.com and scroll to the product I want and one-click buy it.

We tried Second Life, Playstation Home, etc and we didn't like them. Adding whiz-bang 3D isn't going to help a largely failed and discredited concept. The meta-verse is just annoyingly tedious and time-wasting. Those 80s visionaries had no conception of what things like the web or mobile could be. They didn't realize how powerful those technologies are (Hi, we're using it right now to communicate instead of Second Life for a reason). So they projected this idea of "just put people in computers" which helps a dramatic narrative as its easy to understand but isn't very practical in real life.


You're not wrong, but you seem to be missing a huge part of the picture.

Mounting a big black box sucks, yes, but the tech will mature, miniaturize and become more appealing. Even before then, I think many a CEO would rather put on a black box to have a meeting rather than fly for 16 hours.

It's not all about putting meatspace conventions into virtual spaces. The appeal of the metaverse isn't about having to walk around a digital world to get anything done. It's about being able to connect with people across vast physical spaces with a lot of the same quality of the connection as in-person. And more generally, about providing that quality of experience in a ton of contexts.

When buying something online, I think a lot of people would value being able to look at the product from various angles before buying. I know I've cancelled orders because I wasn't sure how a shoe would look based on the one or two angles a site provided. And yes, obviously, you will probably still buy toilet paper and books on amazon just by scrolling in your browser and clicking. But other purchasing processes will be hugely enhanced by VR.

You could better gauge the fit (and when graphics improve, texture/fabric) of clothing. IKEA can go to the next level and actually project the furniture you want into a 3d mockup of the room you want it to go in. Hell, you could probably fly a special drone around your room and it would create the model for you.

Yes, Second Life wasn't everything people imagined. But it was a crude first stab at the problem. Even Oculus is still relatively crude, but it's very immersive already and I don't see why it won't eventually become very refined and convincing. Look at the 3D rendering we're capable of already in games and movies, throw in some Moore's Law and wait to see what comes out.


I'm still not sold on the whole VR idea myself and I loved (and still love) Second Life. Frankly, if I want a share space online I'll just boot up Minecraft or Facebook. They're both just as fulfilling as a full VR headset and take up less mental and physical space.

The only way I see VR catching on is if some energy crisis situation like in Ready Player One occurs. Then there's no other option but to use VR to communicate and do business. And if that situation comes I'll just check out of life (sorry if that's a grim thing to state, but who wants to live through a Hell where you can't even get food?).


Have... have you actually tried it? Have you actually been "inside" a good demo? You stop thinking about things in terms of looking at a screen. You start thinking about things in terms of it surrounding you.

If you have and you didn't get that feeling, that's fine. The tech is still pretty primitive and some people are more readily able to accept it than others (there's a good bit of research recently that shows there is even a hormonal influence). But it's not just a graphics gimmick. It's really a completely different way of thinking about applications.

To me, it literally feels like I'm wearing the program. Not the display. The program. Once exploited properly (and I admit, we're fairly far away from that), we will be doing things in it that will make 2D, external displays too unbearably difficult to use.

I mean, at the very least, we're going to see a sea-change in how 3D content is authored. It's going to massively drive down the cost of creating high-quality 3D models. It's going to make it more accessible, which means it's going to be used more often for things other than just games.

I personally think stereo displays and hand-tracking are a necessary developments to enable in-the-home 3D printing. 3D modeling on a 2D display is too hard for most people to do for even trivial tasks. But we will eventually see a decent modeling program created for an HMD. SketchUp was close, this is going to push it over the edge by making the manipulations far more intuitive.

The closest analogy I have is the difference of going from a text-based adventure interface to a 3D, first-person RPG interface. This is the next step after that. I've gone back and played Skyrim in the Oculus Rift, a game I've spent countless hours in, and been re-amazed at the chance to see such familiar places in literally a whole new way.

Minecraft-style graphics on an HMD feel more real, feel more like being there, than Crysis 2 on the highest settings on an 2D display, more real than watching an IMAX 3D movie. Having experienced a few "tourism" apps, the feeling is closer on the spectrum of reality to that of being there than that of watching it on TV. And that was a year ago on a Google Cardboard. It's only getting better, and it's getting better very quickly. That's what we're dealing with.


If you think scrolling Facebook is as fulfilling as VR then we're on totally different wavelengths (which is totally okay) and I don't know there's much I can say to convince you otherwise.

I do think there are enough people who will love it — for business and pleasure — that it will catch like wildfire, though I wouldn't be surprised if some people only use VR because all their friends do. Just like why I use Facebook.


Minecraft with oculus rift would be amazing.


Slight correction, it is amazing. https://share.oculus.com/app/minecrift


But it would freak me out especially with the Creeper. So, no thanks. I already hate that thing enough.


> It's about being able to connect with people across vast physical spaces with a lot of the same quality of the connection as in-person.

This can be done trivially today with Facetime and other video software. Its largely seen as a gimmick. In the end, people value efficiency, not meatspace equivalents of things. That's why Amazon has all its products with an easy search box and a one click button, instead of a virtual shop assistant.

>I wasn't sure how a shoe would look based on the one or two angles a site provided.

That is a fairly obscure edge case here. Neurotic shoe buyers aren't going to buy a $500 headset and a $1000 gaming PC to buy shoes. They'll take themselves to the store, to you know, actually try the fit, which is fairly important.

>IKEA can go to the next level and actually project the furniture you want into a 3d mockup of the room you want it to go in.

That sounds a lot more like AR, not VR, nor the meta-verse.

>You're not wrong, but you seem to be missing a huge part of the picture.

You sound exactly like the people who were telling me Second Life would change everything about 8 years ago. The concepts are exactly the same, you're just putting on a fancier display. The metaverse is a failed concept for what are fairly obvious reasons. SL proved the skeptics right, so why are we still having this ridiculous conversation?


Apologies in advance for the format and poor writing, I'm very tired!

> This can be done trivially today with Facetime and other video software. Its largely seen as a gimmick.

I find Facetime awkward and not at all like being with someone. That's in contrast to my visit to AltspaceVR a couple of months back. The community relations guy was giving a demonstration of body/handtracking, so his avatar was fairly well articulated. Despite everyone having robotic heads and the rest avatars being simplistic, I very much felt like I was in a courtyard with a group of people watching a man give a talk.

> In the end, people value efficiency, not meatspace equivalents of things. That's why Amazon has all its products with an easy search box and a one click button, instead of a virtual shop assistant.

People (will) value both. It's not like AmazonVR is going to be a giant mall where you walk to the electronics department and browse things by sight. The search box is a very good way to find what you want.

> That is a fairly obscure edge case here. Neurotic shoe buyers aren't going to buy a $500 headset and a $1000 gaming PC to buy shoes. They'll take themselves to the store, to you know, actually try the fit, which is fairly important.

I think GP is talking about a future where VR is cheap and ubiquitous.

> That sounds a lot more like AR, not VR, nor the meta-verse.

Technically augmented virtuality.

> You sound exactly like the people who were telling me Second Life would change everything about 8 years ago.

I never bought into the Second Life hype. Navigating 3d space with 2d devices is never going to feel right outside of a few experiences -- namely games.

> The concepts are exactly the same, you're just putting on a fancier display. The metaverse is a failed concept for what are fairly obvious reasons. SL proved the skeptics right, so why are we still having this ridiculous conversation?

One metaverse as envisaged may very well be a failed concept. That said, the fact that you use the phrase "putting on a display" makes me think you've probably never felt "presence" in VR. It is mind blowing, and that's the reason the conversation is circling back.


acous covered some of my responses, but I still have some stuff to add:

Holding up a phone or looking at a computer screen (or do you look at the webcam??) is nothing like feeling like you're sitting in the same room as your buddies who are themselves in New York and Toronto, while you're all playing a game on a virtual screen together, or just hanging out talking. I don't value efficiency when I want to chat, draw, beatbox or play with my friends.

Buying a shoe is not an obscure edge case. I'm not talking about shoe buyers blowing up the Oculus market when the first release comes out next year. Game players and nerds will handle that. But as acous points out, once the tech is cheaper and more ubiquitous it will take hold in other markets. Have you heard of Zappos? People buy lots of shoes online. They'll buy a lot more when they can play virtual dress-up and match things with outfits.

I think AR & VR will be so intertwined and cross-pollinated that there won't be much point in distinguishing between them pretty soon. But the IKEA furniture planning idea I mentioned could be extrapolated to "pure" VR and a metaverse-type place. I know I would much rather "go to IKEA" in VR than face the gauntlet that is their store-maze. Their whole catalogue will be a VR space you can explore. And I'm pretty confident that it WILL be - they already have super high resolution 3d models of every single product which they use to generate the images for their catalogues. They stopped using physical cameras a while ago.

You sound exactly like the people from 1970 who said "nobody will ever have a computer in their home". I bet Amazon would have sounded like a shit idea in 1985, I'm sure it had tons of naysayers even in the nineties. Same idea here only the timeline will be even more condensed. It's not just a fancier display. It's an entirely different experience. Why are you holding up an 8-year-old game as evidence when technology has come so far in that time?


>> Mounting a big black box on your face is limiting and unappealing to most.

Honestly, everyone says that, until they try it with a good demo, and then they stop caring about the box. The hurdle is getting people to try it. Once they do, if they aren't the rare folks who are super-sensitive to motion sickness, they are immediately hooked.

>> We tried Second Life, Playstation Home, etc and we didn't like them.

Who exactly is "we"? Second Life is not so popular today, but it was pretty huge with some people for a very long time. The fact that it is still running at all after 12 years is surely a testament to its popularity with some people.

And I'm not going to use anything Sony makes as invalidation of any concept.


> The fact that it is still running at all after 12 years is surely a testament to its popularity with some people.

I think the fact that is still running is more of a testament to how much they convinced large corporations and universities to spend enormous amount of money on virtual land, rather than it's popularity or usability among end-users.


If you say so. I've never used it. The only people I now who are using it have been using it forever and are still super psyched about it, and they're all artist-type people, not corporate-shill-type people.


I don't think there's any corporate or university owned sims in SL these days.


>Who exactly is "we"? Second Life is not so popular today, but it was pretty huge with some people for a very long time.

Second Life was a real life and good implementation of the sci-fi metaverse concept. You could trivially build things, script things, monetize things, etc. It was shoved down everyone's throat by marketers and corporate fad chasers looking for the next big thing to reach customers with. It was universally panned for being...well stupid. Meatspace equivalents make zero sense, 3D goggles or not. Plan 9 from Outer Space is a bad movie in both 3D and 2D.

>And I'm not going to use anything Sony makes as invalidation of any concept.

Ironically, Home was a much more stable and mature Metaverse and it targeted gamers exclusively, who would be much more open to the concept as they spend so much time in virtual worlds anyway. Even that failed. No one wants this crap. 3D Rift will be great for AAA shooters. It'll be yet another expensive gaming toy, not a revolution. The same way the Ouya didn't kill consoles and how Linux didn't kill Windows desktop. The nerd friendly PR that Oculus and Facebook have paid for is very much working.


Uhhh, shooters are one of the areas VR is weakest. As it currently stands, it disconnects the visual and physical sensations of moving too much. New technology might change that, but as far as we know, right now, the most successful apps won't be shooters.

I don't know what to say. You clearly have a bug up your ass about it. I doubt I'm going to convince you. I've been there, I've seen it, I work in and on it. I'm saying it's more than just a graphics gimmick. At the very least, it's a completely different way of thinking about applications. If you've experienced VR for yourself and and still don't see that, I don't know what to tell you.


The immediate non-game application that I thought of when I experienced DK1 is the 'perfect seat' movie theater VR experience. Spin up the 'VR Theater' and you walk in to the best seat with the optimal viewing experience every single time you watch a movie.


>> Who exactly is "we"? Second Life is not so popular today, but it was pretty huge with some people for a very long time.

There were two things that killed Second Life's growth.

1. It was poorly managed in every respect (sim tier prices, user base governance and enforcement of TOS, and the poor quality of the infrastructure).

2. Trolls. Trust me, the trolls were good at running just about everyone off. In fact, I believe I know one troll that shot Phil Rosedale's avatar with a penis gun (what do you expect from SL?). So, even when the Grand Pooba of LL can't avoid the trolls you know you're in trouble.


Indeed. This is our goal.


Awesome, this research project looks good to go! I could definitely see this technology being integrated into VR headsets. The strain gauges could come bundled, and the lower half camera could be built into the bottom, and engineered to fit more flush.

Only potential stopper sounds like it requires lot of computation power to measure the face.


Is it necessary to have the screen/VR features of the Occulus Rift to do the facial expression recognition? Or, is data from something more lightweight, e.g. Google Glass, adequate to do facial recognition?

Looking forward to better virtual meetings, less flying.


The facial recognition is from the Intel RealSense. Read the article.


So when can we have our OASIS accounts and dystopian future?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: