9 comments

[ 2.9 ms ] story [ 30.4 ms ] thread
Skycatch is already doing 3D modeling of huge areas using autonomous robots: https://vimeo.com/81128563
Except this isn't 3D modeling of huge areas. This is a drone that can navigate using no other sensors or input except for the Tango.

It's SLAM in a 3 dimensional space without the expensive and cumbersome lasers.

I suppose tango doesn't work so well in direct sunlight?
skycatch appears to be doing realtime 3d mapping though as well.
I don't think skycatch is doing realtime 3d mapping. Most likely they're using photogrammetry that takes a couple hours to process after the flight.
How does Google's Project Tango work anyway?

What I know, it's not LIDAR, nor Stereo-camera or Kinect 1 (infrared laser combined with a monochrome CMOS).

Does it use time-of-flight camera (like Kinect 2) or structure-from-motion (like MS Photosynth, ETH Zurich app)?

Tango uses this special processor (half as powerful as high-end GPU at 160Watt, but it consumes thousand times less battery power):

  It produces over 1 teraflop of processing power on only 
  a few hundred milliwatts of power
-- http://techcrunch.com/2014/02/20/inside-the-revolutionary-3d...

Edit: someone answered it in another thread (today): https://news.ycombinator.com/item?id=7788881

It is "stereo camera" except, instead of using two cameras at a single instance in time it uses one camera at two (or more) instances in time and uses the shifting position of the camera over time to get the multiple perspectives.

The very hard parts are tracking the camera's movement accurately enough to get valid multiple perspectives and the computer horsepower to turn the multiple images (perspectives) into 3D. Obviously, your 3D accuracy is going to be heavily dependent on the accuracy of your camera position, the resolution of the camera, and the ability to do the math in (near) real time.

One trick I've noticed in demos is that they overlay the image on the 3D model. This is really cool, but your eyes and brain will fool you into thinking the 3D model is has better resolution than it really is. If you just display the 3D model as surfaces, you will better see the limitations of the 3D model.

it's called "structure from motion", I am pretty sure Google uses ETH's code (was open source, now ETH's project website vanished), see link above (near "edit").