Slashdot Mirror
Cyber-Goggles Record and Identify Every Object You See
RemyBR writes "Researchers at the University of Tokyo have developed a smart video goggle system that records everything the wearer looks at, recognizes and assigns names to objects that appear in the video. Advanced programs then go back and create an easily searchable database of the recorded footage. Designed to function as a high-tech memory aid, these 'Cyber Goggles' promise to make the act of losing your keys a thing of the past, according to head researcher professor Tatsuya Harada. 'In a demonstration at the University of Tokyo last week, 60 everyday items -- including a potted begonia, CD, hammer and cellphone -- were programmed into the Cyber Goggle memory. As the demonstrator walked around the room viewing and recording the various objects, the names of the items appeared on the goggle screen. The demonstrator was then able to do a search for the various items and retrieve the corresponding video.'" Add in facial recognition technology and this would make for a great aid at conferences and family reunions.
Ass ass boobs ass ass car ass ass boobs OKG PANTIEZZ!!!!! ass ass boobs.
At the bottom of the
Ze goggles, zey do nossing.
Carousel is a lie!
I, for one, welcome our new uhh...pardon me... *click*, *click* ah, yes..our new goggle-eyed...ummm...just a sec...*click*..overcome, no...*click*...overdon, no ..., *click*, *click* yes, overlords!
Hey, where did everybody go?
How would it recognise everyone's faces when they're all wearing the goggles?
The Mothership
Someone in this overpass, somewhere, is bouncing a laser beam off Hiro's face. It's annoying. Without being too obvious about it, he changes his course slightly, wanders over to a point downwind of a trash fire that's burning in a steel drum. Now he's standing in the middle of a plume of diluted smoke that he can smell but can't quite see.
But the next time the laser darts into his face, it scatters off a million tiny, ashy particulates and reveals itself as a pure geometric line in space, pointing straight back to its source.
It's a gargoyle, standing in the dimness next to a shanty. Just in case he's not already conspicuous enough, he's wearing a suit. Hiro starts walking toward him. Gargoyles represent the embarrassing side of the Central Intelligence Corporation. Instead of using laptops, they wear their computers on their bodies, broken up into separate modules that hang on the waist, on the back, on the headset. They serve as human surveillance devices, recording everything that happens around them. Nothing looks stupider, these getups are the modern-day equivalent of the slide-rule scabbard or the calculator pouch on the belt, marking the user as belonging to a class that is at once above and far below human society. They are a boon to Hiro because they embody the worst stereotype of the CIC stringer. They draw all of the attention. The payoff for this self-imposed ostracism is that you can be in the Metaverse all the time, and gather intelligence all the time.
The CIC brass can't stand these guys because they upload staggering quantities of useless information to the database, on the off chance that some of it will eventually be useful. It's like writing down the license number of every car you see on your way to work each morning, just in case one of them will be involved in a hit-and-run accident. Even the CIC database can only hold so much garbage. So, usually, these habitual gargoyles get kicked out of CIC before too long.
This guy hasn't been kicked out yet. And to judge from the quality of his equipment -- which is very expensive -- he's been at it for a while. So he must be pretty good.
It's actually possible to do 3d object recognition in real time eh? 60 objects you say.. hmm, might need support for about 100 times that before use in an autonomous household robot. Some general recognition of categories might be needed too.. otherwise just my cutlery draw will max out 60 objects.
How we know is more important than what we know.
http://www.imdb.com/title/tt0364343/
Now, just integrate the camera and screen directly into the glasses, and give me a Terminator style head-up display!
Bonus for running ROM dumps of the Apple II O/S on the edges...
Will we all look like this when we wear these goggles?
Keep your eyes to the sky.
If you walked around in public with those things on you definitley would not have to worry about remembering ppl's names/faces as people would probably walk on the other side of the road to avoid you. Don't worry about remembering that girl's name/number because they were both fake also.
Personally.. I think paying attention to what your doing and maybe not being constantly on a cell phone/Ipod would be more effective and less time consuming in remembering WTF you are/were doing.
Cyber-gOOgles...
I almost thought this was an article about government spying on surfers...
Previously: "Linux... Toward the Sunrise..." Now: "Linux... Toward the-- No, now, part of Every Sunrise"
It's Edward's goggles! How cool is that, I totally want a pair :)
std::disclaimer<std::legalese> sig=new std::disclaimer; sig->dump(); delete sig;
The Google Goggles will identify everything you see, plus give you relevant ads while seeing them. "Mmm... burger... On sale now at McDonalds!"
i was going to say these would be handy for picking out ex's from the crowd, but then i realized this is slashdot and the chances of more then one female wanting to mate with any of you is remote.
If you mod me down, I will become more powerful than you can imagine....
It's technology like this that is going to kick Big Brother right in the nads!
I think this would be excellent tech to provide to Alzheimer's sufferers, as long as they could remember they had it! If they could program in people's names and play back the last few times they'd seen the person, do a quick review of what they did before they had their nap, review the route they took to where they currently are, etc. that would have a HUGE impact on Alzheimer's sufferers' lives.
Your clothes..give them to me.
First the keyboards came for my handwriting, reducing it to a scrawl Then the speed dial came for number memory - I used to be able to memorize #s for friends and family, no more then the GPS came for spacial awareness. I just follow the voice now. Now they're coming for my other memories. Pretty soon I won't be able to recognize a begonia without help from a computer.
Nullius in verba
As neat as this sounds, I believe it would damage one's ability to remember. With these goggles, the user will rely on it heavily and the memory will suffer from disuse just like any other skill.
They do...something? Bet no one saw that coming.
When they came for the communists, I said "He's next door. Take him away. Goddam commies."
Her name's Yelena Rossini. Anglo-Russian-Italian. Old Heath Road. Her family are so old money they're prehistoric-riche.
I fucked my editor's niece and she says nothing happened. But I know it did. Because I'm clever.
And because I left my shades on. And my shades' defence system thought all the falling down and rolling around and stuff was an assault.
And what does it do when there's an assault? I'm glad you asked.
It takes PICTURES.
-- Spider Jerusalem, Transmetropolitan, 'Year of the Bastard'
Real Daleks don't climb stairs - they level the building.
This reminds me of the Heads up displays in many RTS and RPG games, where you look at an object and it tells you what it is. I can see more advanced versions of this technology doing exactly that, having applications in almost every field, along with military use. And if the database was made large enough, the ability to look at literally anything and know instantly what it is, and is used for would be downright awesome.
Paging Mr. Gibson, Mr. William Gibson....
..but can they counteract the effects of beer goggles to keep slashdotters from taking home questionable members of the opposite sex on a Friday night?
CowboyNeal, CowboyNeal, 3 Stooges, CowboyNeal, CowboyNeal, pizza delivery, CowboyNeal, CowboyNeal, 3 Stooges, OMG Marx Brothers!!!!! CowboyNeal, CowboyNeal, 3 Stooges.
Add in facial recognition technology and this would make for a great aid at conferences and family reunions.
Not to mention nightclubs. Well I guess edittard nailed that already...
you had me at #!
I've dabbled in object segmentation and recognition, and it's a bitch unless you cheat, e.g. by restricting yourself to pre-assigned objects, making sure your objects are always in high-contrast good-lighting conditions, or accepting lots of false negatives or positives. So I'm kind of doubting it will replace the ol' eye-brain recognition system.
That said, I can see some useful related applications. Imagine a helicopter pilot doing search and rescue work. He sees something worth checking out -- say a tiny smoke plume -- and says "bingo." That's picked up by a computer, which is also monitoring his goggles, so it knows in what direction he's looking, and has done some very basic image analysis so it knows to ignore the canopy struts, Sun, shadows et cetera in the field of view. It then combines this with a GPS locator beacon and a good topo map, and instantly computes and records the exact location (latitude and longitude) of the sight of interest. Could save some lives.
Or imagine an emergency worker on the ground during a big fire. He sees a worrying flare-up. Wearing the goggles, he can just say "Looks like trouble over there!" and the goggles, plus associated GPS device and computer, can instantly transmit to headquarters precisely where there is, even if the guy observing doesn't know himself.
These could come in handy..... Helping me remember who that chick that I keep waking up next to is.
Knowing Google's lust for data collection, the Soviet Union is still alive and well inside the psyche of Sergey Brin....
Read the title as "cyber-googles" and got all worked up for nothing.
Move on. Nothing to see here.
Did anyone else hesitate to clink on a link about Japanese researchers which has a domain name of "pinktentacle"?
Why isn't one of the tags for this article virtuallight?
Sounds like a good first step towards Star Trek-style sensors.
"Freon!" Crat cursed. "Just once I'd like to catch some goggle geek alone, with fritzed sensors and no come-go record, then I'd teach 'em it's not polite to stare!"
(h/t some Brin fanboys)
I can see the fnords!
Check out the picture. Do Japanese scientists always dress like that when they're doing science, or was it just Cosplay Friday?
Shouldn't they be tracking your iris so they can see exactly what you are looking at rather then just the direction you are looking in?
Call me when they make goggles that let you see how ugly the chick is no matter how drunk you are!
Once we figure out that whole "gravity" thing, it should make things that much easier when we finally start having real-world katamari-rolling contests. Not only could the goggles tell you what your katamari consists of, they could also tell you how big the thing is relative to any objects contained within it... just like the game versions do now.
Of course, I expect the first person to do it right will be a mad scientist type with a god complex. (Though, I have to admit, launching a ball of screaming brats the size of Texas would be something I'd probably pay to see, even if it wasn't coerced...)
8==8 Bones 8==8
I had some cyber-contact lenses so I could remember where the hell I put my cyber-goggles.
We hope your rules and wisdom choke you / Now we are one in everlasting peace
Cyber-eyes anyone? I don't know whether to be happy to be the first to reference this or sad that no one else made the leap... Terminator would be cool, but the Shadowrun-style cyber-eyes would be way better.
"Does bouncing count?" - Silk, Magician's Gambit by David Eddings
Put the glasses on!
The Rapture is NOT an exit strategy.
"... smart video goggle system that records everything the wearer looks at, recognizes and assigns names to objects that appear in the video. Advanced programs then go back and create an easily searchable database of the recorded footage."
That pretty much describes how my memory works, except I usually remember things with all 5 senses, not just vision. When I need to find something, I close my eyes and call up the sensory data associated with the last time I saw it. Until recently, I thought that was just how everybody's memory worked. With these goggles, I guess it could be.
"Add in facial recognition technology and this would make for a great aid at conferences and family reunions."
The flip side for me is that I am terrible at putting faces with names. Even something as simple as my co-worker wearing a hat and overcoat caused me to fail to recognize him. I could really use something like this for matching people's names to their faces.
http://www.espgame.org/cgi-bin/login
http://en.wikipedia.org/wiki/ESP_Game
"Good news, everyone!"
I sense the legal dogs of war drooling.
Scene: blind guy in a bar ( no not a tavern, this isn't D&D )
AI Interface: Middle age woman, moderately attractive.
blind guy: More details on attractive.
AI Interface: Brown hair, blue eyes, medium build.
blind guy: Is she a candidate for a romantic encounter?
AI Interface: Calculated value for romantic encounter: "I'd do her!"
blind guy: Are you calculating in my current blood alcohol level?
AI Interface: Yes. Besides your blind, all she has to do is feel pretty.
blind guy: Damn vicarious slashdot reading AI coders!
I only look human.
My mother is a halfling and my dad is an ogre, so that makes me an Ogreling
Sure, by humans.
When we talk of software, "very easily" is something a programmer hears when someone who has never programmed is talking.
Could you hand me the whatchamacallit?
Suggestions:
* Make sure object is in proximity
* Stop posting on Slashdot
* Move out of your parent's basement
* Take off google-goggles so you don't look like such a dork
Did you mean: Boys that will date me ?
...at first, then realized that Cyber and Google are mutually exclusive terms.
This is the kind of far-out technology that WE should be working on, not waiting for other countries to develop it then jump on the me-too bandwagon. Unfortunately we're too busy worrying about patent infringement and letting archaic beliefs get in the way of real scientific and technological progress.
Kudos to the Japanese for working on these things. I just wish we were frontrunners ourselves.
And they said zombies weren't real!
Good to see some progress finally being made after all these decades. Been waiting for this sort of stuff and _more_ for quite a _while_.
.here TLD to make it easier to do DNS location of stuff by physical context - http://jukebox.here/ would be a different jukebox in different places, but the ICANN obviously had better things to do like allow domain tasting, and .info, also tried writing an internet draft but it never became and RFC oh well)
:).
Given that the tech for reading brain patterns is getting more and more viable ( http://www.pcpro.co.uk/news/114843/game-on-with-the-braincontrolled-pc.html ), all you need to do is get a computer to associate thought macros with the pictures (and audio) the computer and you are getting at the same time.
Why try replacing the brain with computers to do perception etc? Augment it[1].
1) Your brain is good at perception and recognition.
2) Computers are fairly good at retrieval and retention.
So what you need is to train yourself to semi-automatically "label" "important" stuff with identifiable enough brain patterns/thought macros.
Recognition: you look at stuff, and you might rapidly think "YYY,Picture=Ball,ZZZ" (replace YYY with whatever unique thought macro you choose to tell the computer you're thinking to it, and ZZZ = done").
Retrieval: you could think "Ball", and the computer could show stuff (starting with candidates in a likely context) and you could think "Nope" and then it cycles rapidly (playback) till your brain goes "Yep!" - which would be quite quick after some practice.
If you can't really remember the exact tag, you could try via time ("recent past", etc), or other contexts.
Oh yeah, once you have that thought macro thingy, and add wireless networking, you could think "lights on" and your computer could do it for you - virtual telekinesis (I tried to get the ICANN etc to reserve a
You could also do some virtual telepathy - think of something, retrieve it, share it with someone.
And that's why I think copyright is a big barrier to progress - such pervasive recording and sharing will upset the RIAA etc.
They will want more than a penny for your thoughts, and they'll think those thoughts belong to them too .
And you might be forced to have DRM in your ebrain/iBrain or whatever.
I suggest we better fix that copyright and patent crap now
[1] I also suggest that there are differences between replacing the brain with AI stuff, and augmenting the brain with AI stuff. One might think they seem the same, but the philosophy and goal is different. The long term end result might be the difference between who is the rider and who is the horse.
Oh yeah, this is going to help me tremendously. Now instead of having to try to pick out my boss from all those other passengers, I have to try to pick him out from all the labels blocking my view.
for identifying car parts for production. its very well developed. you look through the goggles and it tells your part number based on analysis if the image.
Is training software to recognize objects, or using them for AI verification, to improve AI for robots. Have a simple validation method, button or voice prompt, tied to the recognition that lets it learn, wear them around for a few months and watch your AI system grow exponentially "smarter." Or at least better trained. Once you have a system that agrees with the average human's perception of their environment, your robot's got it made.
Cybernetic systems like these goggles will raise interesting legal issues. The goggles can record audio. But in states like Pennsylvania and Michigan, it is often illegal to record voice conversations without the prior consent of all parties.
Benjamin Wright, Dallas, Texas, benjaminwright.us