Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Apple Vision predictive bio-feedback

twitter.com

Post reply on HN
Print view View original

141–150 of 192 posts

Re: Apple Vision predictive bio-feedback

#141

Post by viraptor »

OJFord wrote: post #46
viraptor wrote: post #38

Earlier quoted context omitted.

> could be interesting if as users we got access to all these measurements You can! It's not trivial, but not very hard either. eye tracking - OS projects getting this from a laptop camera exist and work well enough for assistive tech (i.e. it can replace your mouse today) electrical activity in the brain - there are fancy things like https://zeto-inc.com/device/ but there are basic options as well heart beats and rh…

I assume GP means that the measurements would be exposed (at some helpful level of abstraction, plus also the raw data say) by the Apple Vision.

I know ; ) I wanted to show there are already alternatives for whose who are interested in experimenting with that data themselves (and at a fraction of the cost)

Re: Apple Vision predictive bio-feedback

#142

Post by 9dev »

sterlingcrispin wrote: post #121

Hey I appreciate the general lack of cruelty so far in the comments, sometimes HN can be pretty rough. The weird phrasing in the tweet is because I can't talk about any of the work I did due to NDA's besides my job description and those patents. But I had a pretty unusual role all things considered. I was also trying to clarify in the tweet that these aren't meant to be interpreted as unannounced existing features or…

I for one got the 10% phrase like you seem to have intended it, and it actually got me to think about how much time I spent in previous employments relative to the entire span of my life, which was an interesting venue to go down, so thank you! (Also for the insights on your work, that sounds like an immensely fascinating experience!)

Re: Apple Vision predictive bio-feedback

#143

Post by makeitdouble »

simontheowl wrote: post #81
makeitdouble wrote: post #7

This whole tweet had the same tone as when pop science crime investigation TV explains a criminal can be identified through its gait alone. It's always magic, except when the predictive model doesn't work and there is no corrective action possible on the user side. We're talking about physical reactions, so that means it's supposed to be the same through gender, age, race, health conditions from the algorithm's point…

A lot of these large companies spend millions of dollars recruiting participants from various demographics to test prototypes and fine-tune technologies like this. It's a tremendous effort.

Yes. Although it's damn hard to recruit a Spanish stay at home mom with a slightly twitching left eye who's too busy to participate to random studies and will receive this a gift for her 50th birthday.

It kinda makes sense that some of her traits would be covered inside the test pannel, but looking at the face unlock and the level of adjustments they're still doing, it doesn't look like they got nearly enough data before launch. And that's for a product that is selling in billions of units.

Re: Apple Vision predictive bio-feedback

#144

Post by makeitdouble »

matt-attack wrote: post #54
makeitdouble wrote: post #25

Earlier quoted context omitted.

You're right. The same way face unlock doesn't work for some people and they inout a password every time, so it's not intractable. I expect people for who eye tracking fails to be stuck on third party controllers for the default interface (hopefully they're supported outside of games ?). The bummer though will be that this device is US only for its launch, so the first round of application might also not care at all…

Well Apple is only expecting less than 1M units to be sold. Must such a niche product be designed for every variation of human? Seems reasonable that if you fall outside of the most common human forms by enough standard deviations then you're just out of luck.

I think people can understand if you tell them their ear is a bit odd shaped, or their makeup throws the dectection off. But it will be weird to explain them their eye movement is really out of norm, especially after they paid 3500.

Re: Apple Vision predictive bio-feedback

#145

Post by furyofantares »

sterlingcrispin wrote: post #121

Hey I appreciate the general lack of cruelty so far in the comments, sometimes HN can be pretty rough. The weird phrasing in the tweet is because I can't talk about any of the work I did due to NDA's besides my job description and those patents. But I had a pretty unusual role all things considered. I was also trying to clarify in the tweet that these aren't meant to be interpreted as unannounced existing features or…

I've used the same 10% metric before. It's funny and I like it.

Re: Apple Vision predictive bio-feedback

#146

Post by justinator »

bdowling wrote: post #139
justinator wrote: post #88

Earlier quoted context omitted.

As is written, the subject of what's, "carefully written" are, "the details [...] in patents".

The "in patents" modifies "available", not "details". Had he wanted to state that the patents were carefully written, he would have written "which", not "and".

Again, original sentence,

"All of these details are publicly available in patents, and were carefully written to not leak anything."

I'll simplify:

"Details available in patents, carefully written to not leak"

What's been written? The patent.

What is in the patent? Details.

How are the details written to not leak? Carefully.

Re: Apple Vision predictive bio-feedback

#147

Post by KaiserPro »

lumb63 wrote: post #124

This begs the question: what is Apple’s real play with Apple Vision? It sounds like there’s a real possibility that, even at a $3500 price point for hardware, the consumer is still the product. What else can their system determine about someone’s brain? What is going to be done with that data?

The eye tracking required to get that level of accuracy is really high. Normally you need a bunch of camera like sensors that operate different to rolling/global shutter, plus a hefty calibration routine to get a good map of the eyeball.

Even then it only really works looking dead ahead with your eye wide open.

so for the moment, that stuff is firmly in the lab.

So whats the real play?

to replace phones with glasses that know you better than you know your self.

To bake all that you see and do into Imagine a really really good personal assistant, that could answer 99.5% of the boring questions asked you with authority. That is the level of AI they (and google and facebook) want

Re: Apple Vision predictive bio-feedback

#148

Post by paul7986 »

sterlingcrispin wrote: post #121

Hey I appreciate the general lack of cruelty so far in the comments, sometimes HN can be pretty rough. The weird phrasing in the tweet is because I can't talk about any of the work I did due to NDA's besides my job description and those patents. But I had a pretty unusual role all things considered. I was also trying to clarify in the tweet that these aren't meant to be interpreted as unannounced existing features or…

Love to start building apps for it like Apple Score (headset keeps score of your real life card game, ping pong match, tennis match, etc & displays it in your view).

How to break into it as I have many ideas Id love to see if they are any good. Are there bootcamps to quickly learn development and around other fellow other eager developers? Im a web designer who uses basic code (html/css/jQuery) to design.

Re: Apple Vision predictive bio-feedback

#149

Post by KaiserPro »

jodrellblank wrote: post #56
BasedAnon wrote: post #41

Right... so this thing hooked up to a centralized AI, deciding what people see when, how often they see it, what they can express and how they feel... Do people really want to live in the matrix? I have a visceral reaction to this idea

https://www.apple.com/newsroom/2023/06/introducing-apple-vis... says " Where a user looks stays private while navigating Apple Vision Pro, and eye tracking information is not shared with Apple, third-party apps, or websites. Additionally, data from the camera and other sensors is processed at the system level, so individual apps do not need to see a user’s surroundings to enable spatial experiences. "

I mean, thats patent bullshit.

sorry, but the eye gase vector will will easy to infer by where its rendering stuff, and where the gesture system is triggered.

Sure apps wont get a look in to the shape of the room persay, but again you can infer it when looking for occlusions.

Re: Apple Vision predictive bio-feedback

#150

Post by KaiserPro »

passion__desire wrote: post #115
rickdeckard wrote: post #32

Earlier quoted context omitted.

If you're a billion dollar company, you invest alot in multiple trajectories of the possible future, because you don't want to end up not being ready when the market shifts (like Nokia for example) and being forced to pay royalty to your competitors just to stay in the market. Now imagine you're a trillion dollar company.

How is one company able to hush all its employees and contractors on the details of such a monumental effort? It's like trying to build a pyramid in secret.

It wasn't that well hushed, there were persistent rumours for years. After all it was the running joke that apple were going to launch a VR headset each WWDC.
Post reply on HN