Live data from Hacker News

Juggalos figured out how to beat facial recognition

theoutline.com

61–70 of 137 posts

Re: Juggalos figured out how to beat facial recognition

#61
post #39

Earlier quoted context omitted.

Speaking of that, I already fed in data from a kinect of the infrared region of interest from a LBP cascade. It works to identify the perceptual hash of the depth and contour of the face. It beats all makeup, completely.

I'll need some makeup artists from Star Trek TNG to help me get ready in the mornings.

Qa'pla

Re: Juggalos figured out how to beat facial recognition

#63

Earlier quoted context omitted.

I would imagine there is a pretty strong first amendment case against such laws in the United States. Mask laws have been successfully challenged on those grounds (though sometimes the challenges have been unsuccessful).

Your bank will likely not let you in if you're wearing a hat or sunglasses. "No shoes, no shirt, no service" has also been a thing for as long as convenience stores have been around. Not that far of a stretch to think this will ultimately extend to face paint. If you want true anonymity, just wear a motorcyle helmet everywhere. We still don't know who The Stig is.

Obviously a public square isn't a bank.

Re: Juggalos figured out how to beat facial recognition

#64

Facial recognition is only “beaten” when a person’s normal face can go undetected without looking like a fucking juggalo .

> Facial recognition is only “beaten” when a person’s normal face can go undetected without looking like a fucking juggalo.

I know this is crudely put and thus being downvoted, but there is a kernel of truth to it similar to the XKCD encyption / wrench[0] comic. Sure, you might've beat facial recognition in the lab, but in a real world scenario you're walking around in ridiculous looking clown face paint so you're going to come under immediate scrutiny regardless.

[0] - https://xkcd.com/538/

Re: Juggalos figured out how to beat facial recognition

#65
post #19

Earlier quoted context omitted.

The world is filled with people that figured out things accidentally because they were doing something else. For example, the person who figured out how to heat food with microwaves (and created the first microwave) didn't set out to do so, he discovered that the chocolate bar in his front pocket melted while working on magnetrons. "Figuring something out" doesn't imply concerted research towards that goal, accidenta…

If I had a chocolate bar melt in my front pocket, I would have assumed it was my body heat.

(I can't find the reference right now, but apparently the chocolate bar story is a myth.)

Re: Juggalos figured out how to beat facial recognition

#66
post #22

More specifically: Juggalos figured out how to beat facial recognition tech trained on normal faces and trying to recognize normal faces. While the black and white paint does seem to negate a lot of natural contrast human faces have – and on which the recognition of faces is currently based – there seems a lot of information left to recognize specific faces, especially when considering stereovision or even possible p…

Speaking of that, I already fed in data from a kinect of the infrared region of interest from a LBP cascade. It works to identify the perceptual hash of the depth and contour of the face. It beats all makeup, completely.

That wouldn't work against the traditional anti-face detection technology: bandanas.

Re: Juggalos figured out how to beat facial recognition

#67
post #56

Earlier quoted context omitted.

They're trying to have a neutral face so it doesn't distract from what they're modelling.

But it's not a neutral face. It's a face of disdain and contempt. This is a neutral face: https://cdn.psychologytoday.com/sites/default/files/styles/a... And this is the face of disdain I'm talking about: https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcSGkhmT...

I know this is subjective but I just don’t see “disdain and contempt” (which assumes the model is trying to communicate something about the viewer) in the sample you provided. I see possibly “secretiveness” and “a challenging disposition” (I assume the model is trying to convey something about herself).

Re: Juggalos figured out how to beat facial recognition

#68
post #56

Earlier quoted context omitted.

They're trying to have a neutral face so it doesn't distract from what they're modelling.

But it's not a neutral face. It's a face of disdain and contempt. This is a neutral face: https://cdn.psychologytoday.com/sites/default/files/styles/a... And this is the face of disdain I'm talking about: https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcSGkhmT...

They both look neutral to me?

Re: Juggalos figured out how to beat facial recognition

#70
post #56

Earlier quoted context omitted.

They're trying to have a neutral face so it doesn't distract from what they're modelling.

But it's not a neutral face. It's a face of disdain and contempt. This is a neutral face: https://cdn.psychologytoday.com/sites/default/files/styles/a... And this is the face of disdain I'm talking about: https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcSGkhmT...

I don't see that second picture on the cv dazzle site at all. All the pictures I see on cv are unquestionably neutral in disposition.
Post reply on HN