“We developed a deep neural network that maps the phase and amplitude of WiFi signals to UV coordinates within 24 human regions. The results of the study reveal that our model can estimate the dense pose of multiple subjects, with comparable performance to image-based approaches, by utilizing WiFi signals as the only input.”

    • Mossy Feathers (She/They)@pawb.social
      link
      fedilink
      English
      arrow-up
      35
      arrow-down
      4
      ·
      edit-2
      1 year ago

      Also VR nerds. Current tracking is either based on the headset, so you can’t move your arms unless the headset can see them, or your arms have to be seen by lighthouses, or you rely purely on gyroscope and accelerometers for tracking, which tend to drift. So either you have blind spots, have to deal with occlusion, or will slowly drift and have to recalibrate periodically. Wifi-based tracking seems like a neat idea tbh.

      Edit: considering wifi is just photons that aren’t wiggling fast enough for us to see, I’d be surprised if the government doesn’t already have this technology behind closed doors.

    • assassinatedbyCIA@lemmy.world
      link
      fedilink
      English
      arrow-up
      2
      ·
      1 year ago

      Honestly, I wouldn’t be at all shocked if uncle sam and his favourite three letter agencies have been playing around with this tech for years.

  • Sundray@lemmus.org
    link
    fedilink
    English
    arrow-up
    62
    arrow-down
    1
    ·
    1 year ago

    I mean you can look through my walls if you want, but don’t come crying to me if you don’t like what you see.

    (I’m painting fantasy miniatures. They’re for a friend.)

  • Agent641@lemmy.world
    link
    fedilink
    English
    arrow-up
    45
    arrow-down
    2
    ·
    1 year ago

    You know what else let’s you see through walls? Windows. (Suck it, Linux users!)

    • MNByChoice@midwest.social
      link
      fedilink
      English
      arrow-up
      15
      ·
      1 year ago

      Article is from a year ago. Government tends to be ahead of the curve. As an uninformed guess, they have been using it in high value situations for 4+ years.

      (Dear FBI, the above is a guess based on public information. I don’t know shit.)

      • DdCno1@kbin.social
        link
        fedilink
        arrow-up
        9
        ·
        1 year ago

        Israel has been using a similar system since at least 2022:

        https://petapixel.com/2022/06/29/the-xaver-1000-sees-through-walls-and-is-made-for-the-israeli-army/

        It’s pretty likely that they have shared this system with their closest allies, similar to how the Trophy missile defense system found its way onto German and American tanks.

        By the way, those throwable cameras mentioned at the end of the article have been available to the IDF since 2005.

      • abhibeckert@lemmy.world
        link
        fedilink
        English
        arrow-up
        10
        arrow-down
        1
        ·
        1 year ago

        Government tends to be ahead of the curve.

        I dunno what world you’re living in, but I live in a world where police still do nearly all their work with pencil and paper and if you want to talk to a police officer, no you can’t talk to them on the phone or send an email. You’ll have to have a meeting face to face.

        • Cethin@lemmy.zip
          link
          fedilink
          English
          arrow-up
          8
          ·
          1 year ago

          Actually, police do tend to be reasonably ahead when technology is invasive. I’ve heard many stories about them seeing through walls with other technology. They also tend to like face-tracking, which is pretty advanced. They just are really far behind on technology that could be used to hold them accountable.

  • xia@lemmy.sdf.org
    link
    fedilink
    English
    arrow-up
    19
    ·
    1 year ago

    Which leads to the obvious question: how long has the military been able to do this?

  • Atelopus-zeteki@kbin.run
    link
    fedilink
    arrow-up
    17
    ·
    1 year ago

    Umm, article from Jan 19, 2023. I remember seeing it then. Is there anything new on this?

    The paper: [Submitted on 31 Dec 2022] - I’m not sure if it’s out of prepublication yet.
    DensePose From WiFi - https://arxiv.org/abs/2301.00250

    Here’s another tidbit from July 2, 2023 DensePose from WiFi - See through the walls. - https://www.linkedin.com/pulse/densepose-from-wifi-see-through-walls-alejandro-fernández

    And another from July 25, 2023 Revolutionary Applications of DensePose From WiFi: Enhancing Corporate Security and Empowering Military Tactical Teams - https://www.linkedin.com/pulse/revolutionary-applications-densepose-from-wifi-enhancing-zack-hamm

  • Maggoty@lemmy.world
    link
    fedilink
    English
    arrow-up
    18
    arrow-down
    3
    ·
    1 year ago

    Duh? I don’t think anyone with the right field of study thought this wasn’t possible. It just doesn’t have good use cases.

    • Milkyway@feddit.de
      link
      fedilink
      English
      arrow-up
      6
      ·
      1 year ago

      I’m an EE, and I have serious doubt about this actually working nearly as good as they are putting it. This sort of stuff is hard, even with purpose built radar systems. I’m working with angle estimation in Multipath environments, and that shit fucks your signals up. This may work it you have extremely precisely characterised the target room and walls, and a ton of stuff around it, and then don’t change anything but the motion of the people. But that’s not practical.

      • Maggoty@lemmy.world
        link
        fedilink
        English
        arrow-up
        3
        ·
        1 year ago

        It’s Popular Mechanics, of course it doesn’t work as well as they say it does. But the theory has been around a long time.

      • Socsa@sh.itjust.works
        link
        fedilink
        English
        arrow-up
        2
        ·
        1 year ago

        You are correct, at best this requires some a priori knowledge of the room. You can kind of do basic motion detection blindly though. They are just measuring the channel response via the 802.11 preambles, so for basic presence detection knowing that the channel response is changing is enough.

      • Nommer@sh.itjust.works
        link
        fedilink
        English
        arrow-up
        1
        ·
        1 year ago

        I was under the impression these experiments required a pre mapped room with EM readings. I don’t think they can watch you like if it was an X-ray but I’d believe it if they could track blobs of moving mass.

  • Xavier@lemmy.ca
    link
    fedilink
    English
    arrow-up
    17
    arrow-down
    2
    ·
    1 year ago

    Henceforth, the building code shall make mandatory that every room be perfectly grounded Faraday cages (/s).

    Still, imagine lethal drones integrated with that technology (of course, they already have infrared, maybe even some adequate wavelength of X-rays).

    Nevertheless, pretty cool to see how far we can take preexisting technology with the help of some deep learning layers.

    • Maggoty@lemmy.world
      link
      fedilink
      English
      arrow-up
      7
      ·
      1 year ago

      Here’s what they’re putting in the goggles that Infantrymen wear now.

      I don’t care to guess what the drones are packing.

      • Shurimal@kbin.social
        link
        fedilink
        arrow-up
        3
        arrow-down
        1
        ·
        1 year ago

        What we know about drones is that they have cameras that can discern individuals from 10 km altitude.

        What we suspect is that US has Hubble-sized spy satellites that can do almost the same. There were a lot of classified military STS missions.

        What is theoretically possible is that US drones and spy sats can function as very large arrays (we do this with astronomical telescopes already) to dramatically increase spatial resolution.

        • Maggoty@lemmy.world
          link
          fedilink
          English
          arrow-up
          2
          ·
          1 year ago

          I’d believe it. When I was in the infantry 20 years ago we could see you 3km away with the optics mounted on our machineguns. And several kilometers with cameras mounted on towers. I don’t know how far they went but it was at least 5km because we were directing mortar fire with them and that’s about the range of the mortar system we were using.

        • SeducingCamel@lemm.ee
          link
          fedilink
          English
          arrow-up
          0
          ·
          1 year ago

          Oh I wonder if that’s how the Pic was taken that trump tweeted out of that rocket launch site, people didn’t think it was physically possible for a satellite to have that resolution

          • Shurimal@kbin.social
            link
            fedilink
            arrow-up
            1
            ·
            1 year ago

            It all comes down to the size of the mirror/lense—the bigger, the better. Up to a point. The biggest problem is air currents and different air densities refracting light and distorting the image. That’s what these laser beams are for on photos taken of astronomical observatories—they give reference light spot that can be used to calibrate adaptive optics to current atmospheric conditions reducing distortion.

    • theodewere@kbin.social
      link
      fedilink
      arrow-up
      1
      ·
      1 year ago

      near future building insulation will include enhanced EM spectrum insulation, to prevent or distort leakage into the environment

    • Chozo@kbin.social
      link
      fedilink
      arrow-up
      14
      ·
      1 year ago

      The tech is even far older than the article. I remember seeing this being demo’d at least a decade ago. Though, it looks like the fidelity has improved significantly from the early proof-of-concept videos that were floating around for a while.

    • gedaliyah@lemmy.worldOP
      link
      fedilink
      English
      arrow-up
      4
      arrow-down
      1
      ·
      1 year ago

      Sorry, I didn’t notice the date when I posted. I can take it down if requested.