Modern day image recognition is getting pretty sophisticated, such that people can be easily identified by cameras and automated systems. If I recall correctly, even the way a person walks can help ID them.

(The reason I'm asking this at all is because of my limited understanding of this tech, the uniqueness of human mannerisms, and how the two are used together.)

Given a current or very near-future level of surveillance tech, would it even be possible to train a small group of people (let's say 5) to move and act similarly enough that automated systems would not be able to tell them apart (including non-real-time analysis)?

Assume that the individuals could be wearing a mask of some kind to attempt to foil facial recognition, and are carefully selected or trained to be as physically similar to each other as possible.

Edit for Clarification: I'm referring to things like video cameras and image recognition/analysis, not things like biometric scanners. I thought that was sufficiently clear already, since I mentioned "image recognition", "surveillance", and "facial recognition", but apparently some people weren't 100% positive.

  • 3
    $\begingroup$ Do you mean if a computer followed each one all day long every day or during a dance routine or in what circumstance is the machine observing them ? $\endgroup$
    – Raditz_35
    Oct 12, 2017 at 2:27
  • 1
    $\begingroup$ put on a pair of glasses and change them regularly, don't worry about gesture I have been trying to teach my Kinect to recognize me... $\endgroup$
    – user6760
    Oct 12, 2017 at 2:39
  • 1
    $\begingroup$ What do you mean with "modern day recognition"? Facial recognition? Iris scanner? DNA sequencing? $\endgroup$
    – L.Dutch
    Oct 12, 2017 at 2:50
  • $\begingroup$ This is greatly depends on how the computer tells them apart. A facial recognition can be fooled even by a photo. $\endgroup$
    – Vylix
    Oct 12, 2017 at 5:32
  • 1
    $\begingroup$ If you want to get a feel for how easy it can be to fool machine vision systems have a look at this - particularly the bit about making a panda "look" like a gibbon $\endgroup$
    – Samwise
    Oct 12, 2017 at 21:45

3 Answers 3


I'm going to make a HUGE assumption: you're talking about the kind of software that would track people in the streets. Not software that would have the chance to closely examine people like access scanners. If you want to fool every scanner on earth, the answer is "absolutely not." I can't think of anything you can do to fool a retina scanner without cloning the target... and even that might not work.

Given the above assumption — and remembering that no recognition system is perfect...

The highest-end recognition systems are still only able to give you a statistically-correct answer (e.g., "92% match..."). This works tremendously in your favor. Especialy where mannerisms, gait, etc are involved because almost anything can cause those to change. "Correct" may be something closer to 60% (don't believe me? Sit in front of a TV for 3 hours, then walk in front of the scanner and see if it recognizes you with the limp from a sleeping leg. Or a twisted ankle. Or the bruise on your arm from trimming the tree....)

And I've seen makeup turn one person into another so completely that photographically a human can't tell the two apart. Remember, computers are far from perfect. Optical resolution, angle of scan, clouds in the sky, light reflecting off irises, not to mention motion. The engineer within is laughing hysterically over the idea that exclusion today could ever be better than 90% (and that might be wishful thinking).

Therefore, absolutely. Obviously, the exact circumstances (which you didn't provide) can affect my answer, but with no other data, yup, it can be done.

What are the actual points of potential failure?

  • eye separation.

That's it. Remember, you're dealing with a computer and an imager. No matter what people tell you, you can make bone structure and everything else look to be someplace completely different just using makeup. And if the software is using pupil location to justify "eye separation" then it could be fooled with contact lenses that adjust the "location" of the pupil. It it's using eyesocket or open-eye imaging, then you're back to fooling it with makeup. You might look a bit funny to a human, but the computer would be completely fooled.

Anybody who tells you differently is trying to sell you the system.

So, why don't people do this to fool the systems all the time? Because for nearly nine billion of us, it simply doesn't matter, and so we don't care. That works in the favor of people running the software. If half the people in a city starting acting to intentionally foil such systems (wearing hats, growing beards, wearing sunglasses, sticking cotton balls under your lips...) the usefulness of the software would drop to almost zero.

Unless we're talking about access scanners. Remember what I said in the beginning, it's much harder to fool access scanners because light, motion, angle, etc. are all controlled. That increases the liklihood of a good scan by the proverbial orders of magnitude.


I'm assuming that the technology is the same as today (2017). If so, then the answer is no. If a computer has a large enough database to identify someone out of let's say, millions, then no. No matter how much training you put into someone to impersonate another person, that trainee will still have a little bit of their own characteristics left, and that's just the obvious physical characteristics and features. Even with plastic surgery one cannot be identical to another person.

Like with your example, even how a person walks can identify them. No matter how much training you put into a person's gait, it wouldn't be enough to be second nature to them. Iris scanners can identify you with the unique patterns that are in your irises, much like a fingerprint.

As for the mask, it depends on the material it's made of. If it's a digital mask, then there might be a chance that the computer won't recognize it, but if it's an actual mask, unless it's made of the skin of the person your agent is trying to impersonate, then no, computers won't be tricked by it.

  • $\begingroup$ I think the point of the mask would be that all 5 would wear the same mask, taking facial recognition off the table. Otherwise, unfortunate but solid answer. $\endgroup$
    – 11684
    Oct 12, 2017 at 7:25
  • 1
    $\begingroup$ Imho your answer is just plain wrong. the first part (large enough database) will actually make it worse for the computer. Say you have a database of 10 people, then a 90%match will identify one person very likely. make that 10'000 people, and a 90% match leaves you with 1'000 people who might be the subject. And about the gait: different clothes already affect your gait, shoes in particular. and then there is training. Actors do that all the time. loose clothing can make the details harder to identify. $\endgroup$
    – Burki
    Oct 12, 2017 at 8:11

It depends. The facial recognition software on a Samsung Galaxy S8 can be tricked by a photo, and (in the US, at least) you aren't allowed to smile for your driver's license, since that screws up facial recognition.



However, the software also uses "the relative position, size, and/or shape of the eyes, nose, cheekbones, and jaw", and there's no training which can overcome that".



You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .