4 ms·

The thing is, they won’t work forever or even for very long. Even makeup will eventually be thwarted.

quickthrowman 7y ago
I remember seeing an article about how ICP “juggalo” makeup fools some facial recognition software, I wonder if that has been solved yet?
- Igelau 7y ago
  The juggalos are evolving. They have superior technology. We thought we'd finally cracked the code, but it was too late: the enemy was already within the gates. Our hopes were doused in a shower of Faygo, and the machine lay sizzling its last breath in the putrid foam.
falcolas 7y ago
CV dazzle has been around for at least 7 years; an eternity in this era of CV growth. It still works. What makes you believe it will be thwarted?
- ceejayoz 7y ago
  As with ad blocking, no one really cares until it becomes widespread. CV doesn't account for dazzle because no one's really using it; there are more immediate problems to solve of higher utility. There's also the non-technical approaches, like https://en.wikipedia.org/wiki/Anti-mask_law https://en.wikipedia.org/wiki/Anti-mask_law.
- freedomben 7y ago
  I'm certainly no expert in CV, but having worked in government/law enforcement tech previously (pre-Snowden days) I have full confidence that if people started actually using CV Dazzle to evade detection, they would work on breaking through it. Generally speaking those systems only receive features that are actually problems the users are facing. Once they identify the problem they can throw nearly unlimited amounts of money at it, and they will win. This is why I'm focusing my efforts on political solutions to protect our privacy, rather than technical solutions. That's not to say I don't think technical solutions are important, because they absolutely are, but even things like encryption could be made illegal politically if they want to. Thus politics is the most important arena for privacy advocates IMHO.
  - lozaning 7y ago
    And the issue now is that as much as you might blend in to the computer, if you're going out and about with CV dazzle on you're going to attract a lot of scrutiny from actual people. Like this: https://xkcd.com/1105/ https://xkcd.com/1105/
  - AnthonyMouse 7y ago
    The argument that we need political solutions originally came from a time when technology people weren't paying attention to politics at all, and of course then you get this: https://xkcd.com/538/ https://xkcd.com/538/ But it feels like it has gone the other way now, to the point that people dismiss potentially helpful technical solutions because "what really matters is politics" even though in reality you need both. If the government bans privacy technologies then it's hard to use them, but if nobody develops or uses them to begin with then you don't have them either. And for some people the government in question isn't ours, it's a foreign authoritarian government where "policy solutions" are just not going to happen and the only option available is technology. It's a lot easier to develop that technology in less authoritarian countries, which means we have a responsibility to do it, because if they have to do it the cost of a mistake is a lot higher.
    - freedomben 7y ago
      Agree with you completely. If I take my previous statement further to "don't even bother holing up as insurgents because the government will win the battle" then it goes to a place I don't agree with (I think there's plenty of evidence that the "little guy" can win against big governments. Afghanistan is a great example). So I really should re-phrase my position to clarify that I meant in the United States, but also even in the US I do think the technological solutions are important and should be developed. The important point I want to stress is that I don't think we should ever disregard the political war because of technological confidence in our ability to thwart the authoritarians. Both politics and technology are both critical pieces. If we lose on either front, we lose.
- pathseeker 7y ago
  Nobody uses is so it hasn't been anything worth thwarting.
currymj 7y ago
when they're generated digitally, it's surprisingly hard to get good robustness against adversarial examples, and people have been working on it for a while. of course you can train to stop any specific input, but then new ones can be created for the newly-trained network. there are techniques that can prevent this, but you pay a real price in overall performance, which may or may not be worth it. of course it's harder to make physical objects work, but if somebody did manage to make a method for physical adversarial attacks that worked reliably, I would guess you'd run into the same problems. it's not at all clear that the endgame is that we can easily stop these attacks.
- bidart 7y ago
  The issue with adversarial examples is defending against a wide class of examples, not a few specific images. Clothes change much slower than software updates, and defending against a few adversarial examples is relatively easy, so this shouldn't be a problem.
  - Retric 7y ago
    NN updates can take a long time to be trained, verified, and deployed. Especially if your talking a wide scale network of cameras. Some of these approaches are 7+ years old and still work.
    - clort 7y ago
      long term storage overcomes the deficiencies in todays software As soon as they update the software, they run it over the stored video and investigate the differences. When they identify particular evasion techniques, they can flag up all instances and see what was done under their cover.
  - egd 7y ago
    >Clothes change much slower than software updates, I change my clothes at least once a day and add to my wardrobe on a ~weekly to ~monthly basis. Having been party to large system upgrades at even a reasonably competent company, I'd be shocked to find out that any surveillance system anywhere is matching that velocity.

I remember seeing an article about how ICP “juggalo” makeup fools some facial recognition software, I wonder if that has been solved yet?

Igelau 7y ago
The juggalos are evolving. They have superior technology. We thought we'd finally cracked the code, but it was too late: the enemy was already within the gates. Our hopes were doused in a shower of Faygo, and the machine lay sizzling its last breath in the putrid foam.

falcolas 7y ago

CV dazzle has been around for at least 7 years; an eternity in this era of CV growth. It still works. What makes you believe it will be thwarted?

ceejayoz 7y ago
As with ad blocking, no one really cares until it becomes widespread. CV doesn't account for dazzle because no one's really using it; there are more immediate problems to solve of higher utility. There's also the non-technical approaches, like https://en.wikipedia.org/wiki/Anti-mask_law https://en.wikipedia.org/wiki/Anti-mask_law.
freedomben 7y ago
I'm certainly no expert in CV, but having worked in government/law enforcement tech previously (pre-Snowden days) I have full confidence that if people started actually using CV Dazzle to evade detection, they would work on breaking through it. Generally speaking those systems only receive features that are actually problems the users are facing. Once they identify the problem they can throw nearly unlimited amounts of money at it, and they will win. This is why I'm focusing my efforts on political solutions to protect our privacy, rather than technical solutions. That's not to say I don't think technical solutions are important, because they absolutely are, but even things like encryption could be made illegal politically if they want to. Thus politics is the most important arena for privacy advocates IMHO.
- lozaning 7y ago
  And the issue now is that as much as you might blend in to the computer, if you're going out and about with CV dazzle on you're going to attract a lot of scrutiny from actual people. Like this: https://xkcd.com/1105/ https://xkcd.com/1105/
- AnthonyMouse 7y ago
  The argument that we need political solutions originally came from a time when technology people weren't paying attention to politics at all, and of course then you get this: https://xkcd.com/538/ https://xkcd.com/538/ But it feels like it has gone the other way now, to the point that people dismiss potentially helpful technical solutions because "what really matters is politics" even though in reality you need both. If the government bans privacy technologies then it's hard to use them, but if nobody develops or uses them to begin with then you don't have them either. And for some people the government in question isn't ours, it's a foreign authoritarian government where "policy solutions" are just not going to happen and the only option available is technology. It's a lot easier to develop that technology in less authoritarian countries, which means we have a responsibility to do it, because if they have to do it the cost of a mistake is a lot higher.
  - freedomben 7y ago
    Agree with you completely. If I take my previous statement further to "don't even bother holing up as insurgents because the government will win the battle" then it goes to a place I don't agree with (I think there's plenty of evidence that the "little guy" can win against big governments. Afghanistan is a great example). So I really should re-phrase my position to clarify that I meant in the United States, but also even in the US I do think the technological solutions are important and should be developed. The important point I want to stress is that I don't think we should ever disregard the political war because of technological confidence in our ability to thwart the authoritarians. Both politics and technology are both critical pieces. If we lose on either front, we lose.
pathseeker 7y ago
Nobody uses is so it hasn't been anything worth thwarting.

currymj 7y ago

when they're generated digitally, it's surprisingly hard to get good robustness against adversarial examples, and people have been working on it for a while. of course you can train to stop any specific input, but then new ones can be created for the newly-trained network. there are techniques that can prevent this, but you pay a real price in overall performance, which may or may not be worth it. of course it's harder to make physical objects work, but if somebody did manage to make a method for physical adversarial attacks that worked reliably, I would guess you'd run into the same problems. it's not at all clear that the endgame is that we can easily stop these attacks.

bidart 7y ago
The issue with adversarial examples is defending against a wide class of examples, not a few specific images. Clothes change much slower than software updates, and defending against a few adversarial examples is relatively easy, so this shouldn't be a problem.
- Retric 7y ago
  NN updates can take a long time to be trained, verified, and deployed. Especially if your talking a wide scale network of cameras. Some of these approaches are 7+ years old and still work.
  - clort 7y ago
    long term storage overcomes the deficiencies in todays software As soon as they update the software, they run it over the stored video and investigate the differences. When they identify particular evasion techniques, they can flag up all instances and see what was done under their cover.
- egd 7y ago
  >Clothes change much slower than software updates, I change my clothes at least once a day and add to my wardrobe on a ~weekly to ~monthly basis. Having been party to large system upgrades at even a reasonably competent company, I'd be shocked to find out that any surveillance system anywhere is matching that velocity.