5 ms·
People wear facemasks in camera footage now, so you have to go by their pants, belts, shoes, fingerprints..
by bluedino 4y ago
People wear facemasks in camera footage now, so you have to go by their pants, belts, shoes, fingerprints..
- TechBro8615 4y agoGait analysis is a more surveillable fingerprint anyway, since it can work with lower resolution cameras. Tag someone on a high res camera with facial recognition and as long as you've seen their gait once, you'll be able to identify them again.
- jamiek88 4y agoStones in our shoes, masks on our faces, strobing hoodies a boring dystopia indeed.
- TechBro8615 4y agoDiscussed previously, apparently (found via Google search for `anti gait recognition shoes`, lol): https://news.ycombinator.com/item?id=18401892 https://news.ycombinator.com/item?id=18401892 I like the comment about the Ministry of Silly Walks.
- mananaysiempre 4y agoIIRC stones in shoes as a gait recognition defence were a plot point in Doctorow’s Little Brother (2008).
- jamiek88 4y agoYeah I read that in a novel too. It’s a common idea.
- beardedwizard 4y agoBecause spies actively use the technique and it's routinely discussed in that setting.
- MonkeyMalarky 4y agoAn uncomfortable and time consuming one indeed.
- pvarangot 4y agoI don't think gait analysis would hold itself through a trial or be bought by a jury. Maybe to correlate different videos of possibly different suspects? But if you are getting regularly videoed by security cameras while circumventing rules you are probably already wearing different shoes or pants and that can go a long way to defeat gait analysis algos.
- devmor 4y agoIt's quite literally more accurate than fingerprint or blood splatter analysis and those send people to prison on a daily basis - despite being little more than snake oil techniques.
- zirgs 4y agoHere's a handy guide: https://en.wikipedia.org/wiki/The_Ministry_of_Silly_Walks#/media/File:Silly_Walk_Gait.png https://en.wikipedia.org/wiki/The_Ministry_of_Silly_Walks#/m...
- jandrese 4y agoHas gait analysis been studied by people other than the ones selling gait monitoring equipment? It smells to me like bite mark analysis, blood splatter analysis, polygraphs, and other "miracle" technologies/techniques that turned out to be horsecrap.
- mcbits 4y agoThere are people I know that I can recognize by the way they walk, so the idea is at least plausible. I think the main problem is/will be from using the analysis inappropriately. It's one thing to narrow down a city-sized population to a list of 50 people known to walk with a similar gait, and then follow up with other forms of investigation before drawing conclusions. It's another for a cop to look at the top match and go arrest that guy, as they've done more than once with facial recognition.
- anigbrowl 4y agoGait analysis is absurdly over hyped. I wish people would stop trying to win arguments by mentioning at think twice about the numerous failure cases.
- tiagod 4y agoGait analysis also works surprisingly well and I don't see how to make it less effective without damaging your body
- blobbers 4y agoShow me a security camera that does gait analysis and I will pay you $1000.
- aYsY4dDQ2NrcNzA 4y agoYou’re offering to pay a grand to random HN contributors to wire one up.
- dylan604 4y agoMaybe that's the point? Maybe the person needs one of these and $1000 is well worth it, and it just needs to be
- aYsY4dDQ2NrcNzA 4y agoI don’t get that sense.
- dylan604 4y agoThe person knows no cameras exist doing gait detection, but wants one. They make a random comment on HN, a forum known to be read by people that is full of people will take an idea to completion "over a weekend", as a challenge to get someone to do something for $1000. That's a ridiculously low amount for something to come into existence that would serve their need directly. Hell, there's probably people here that would do something like this for the clout.
- blobbers 4y agoNo, I know that where security cameras are positioned and the amount of processing they currently have there will be no gait analysis being done on the camera itself. They have enough trouble telling one person from another reliably. I was being facetious as I don't believe hackernews readership is that enterprising.
- wstuartcl 4y agoThere have been pretty huge leaps over partially obstructed facial recognition algorithms in the last few years -- I think state of the art is approaching no meaningful loss of match mask vs no mask.
- blobbers 4y agoNegative ghost rider. It is most certainly not, unless you’re talking straight on well lit training data.
- wstuartcl 4y agoThis is not my understanding, I have read at least 7 or 8 papers that seem to have for various models and techniques reduced the delta between masked and unmasked recognition to be very similar on false positive and positive rates. https://www.mdpi.com/2076-3417/11/16/7310 https://www.mdpi.com/2076-3417/11/16/7310 https://www.sciencedirect.com/science/article/pii/S2405896322007558 https://www.sciencedirect.com/science/article/pii/S240589632... ... That said I have no insight as to how many of these techniques have been found to scale well or have started to make it into product. It has been publicly reported that NEC’s NeoFace (a system that many police and govt use) newer versions does indeed have occlusion (mask) recognition operating at very high levels. anyways thats just my understanding as an interested bystander -- not in the field.
- snapcaster 4y agoHow could this be true? I don't follow the field but so much of the signal is being obscured it's hard to imagine how accuracy wouldn't suffer
- flangola7 4y agotldr machine learning is magic iPhone face unlock has worked while wearing a mask for a while now. You know how you can identify exactly which family member is walking in the door from across the house, and maybe even what mood they're in? Or how little clarity you need to identify your child or spouse on a 480p camera feed? That's what machine learning makes possible across all types of sensor input, picking out those little but distinct patterns. There's really no way to be anonymous in public once ML surveillance software is widespread.