Leap Motion

» see leapmotion.com as it lived, on the wayback machine

~ the obit ~

leap motion's controller landed in july 2013. by then i had finished my masters and was working in computer vision in ankara. the demo videos came around tech forums that summer and looked impressive, the small usb dongle that sat in front of your keyboard and saw your hands hovering above it. i did not buy one. nobody i worked with did either, for some reason it was always treated as a thing you played with once at a demo booth and then went back to your real workflow.

david holz had started the optical-tracking work in 2008 while doing a phd in mathematics, founded the company with michael buckwald in 2010 as ocular wave (then onyx motion, then leap motion). the technology was two monochromatic infrared cameras and three infrared leds aimed up at the volume of air above the device. submillimeter precision at peak, way past anything microsoft kinect was doing at the same range. the public reaction in 2012-2013 was that it had to be the future of something. it raised $300m at a peak valuation that year, best buy stocked the controller, hp shipped a keyboard with a leap built in.

back then the use case was vague. you could use it as a desktop mouse alternative, except holding your arm up in the air was tiring after about four minutes. you could use it for short demos, except no production software ever adopted it. by mid 2014 leap motion had sold about 500,000 units against a much higher internal forecast and had laid off about 10% of staff. the company was already running out of room.

~ the rap sheet ~

~ the rise ~

the early demos got everywhere. ted talks, fast company covers, the youtube ad-roll. for the early 2013 hype cycle leap motion was the consumer-tech-next-big-thing in the same way kinect had been three years earlier and oculus dk1 was the same year. the small price ($79 retail in the us) and the small footprint made it look like it could end up on every desk by 2016.

the pivot to vr came in february 2016 with the release of the orion software, which retargeted the controller for hand-tracking in vr headsets. the device could be velcroed to the front of an oculus rift or htc vive and would track your hands inside the vr scene. the orion demos were sharp. the vr install base was small. neither the rift nor the vive shipped a year that the leap motion peripheral could attach itself to in volume.

~ the cracks ~

the gesture-input wall is a wall every camera-tracking company has hit since the 1990s. the problem is not the tracking, the tracking can be made good enough. the problem is that pointing at things in the air does not do anything a mouse does not do better, and waving for menu navigation gets tiring fast. every vendor that has shipped good hand-tracking has watched the user base default back to controllers inside a week. kinect for xbox in 2010 and leap motion for desktop in 2013 hit the same wall, the meta quest 2's built-in tracking found it again years later. the controller knows what you want to click.

by 2018 leap motion had pivoted again, this time toward enterprise and automotive cockpit applications. the consumer controller was being deprecated in favor of an embedded module aimed at car interior gesture controls. apple was rumored to be circling for an acquisition in 2018 (the deal never closed). the company was burning cash by then and the round was not coming.

~ what it became ~

ultrahaptics, a british company that made ultrasound-based mid-air haptics, bought leap motion on may 30 2019 for around $30m. the combined entity was renamed ultraleap and refocused on the automotive cockpit and digital-signage gesture-input segments. the desktop usb controller was eol-ed around the same time, the leapmotion.com domain redirected to ultraleap.com.

in 2026 ultraleap is still around as an enterprise hand-tracking and haptics company, mostly invisible to the consumer side it once dominated press cycles in. the original $79 controllers still work on a windows pc with the legacy driver, if you can find a copy of the old sdk. on every retro-tech subreddit there is one post a year asking what happened to leap motion, the answer is that it sold its tech for 10% of its peak valuation and is now a b2b haptics company nobody hears about.

@}-->--

              )
             (_)
            .-'-.
            |   |
            |   |
            |   |
            |   |
          __|   |__   .-.
       .-'  |   |  `-:   :
      :     `---'     :-'
       `-._       _.-'
           '""""""

« previous  ·  random  ·  next »