0

November 14, 2010 AT 8:46 pm

Hacked Kinect is now a 3D video capture tool

Hacked Kinect is now a 3D video capture tool, b00m! This is so cool.


Check out all the Circuit Playground Episodes! Our new kid’s show and subscribe!

Have an amazing project to share? Join the SHOW-AND-TELL every Wednesday night at 7:30pm ET on Google+ Hangouts.

Join us every Wednesday night at 8pm ET for Ask an Engineer!

Learn resistor values with Mho’s Resistance or get the best electronics calculator for engineers “Circuit Playground”Adafruit’s Apps!


Maker Business — Shenzhen: The Silicon Valley of Hardware (Full Documentary) and Who invests in hardware?

Wearables — Take flight with shiny wings

Electronics — Inadequate volt signal

Biohacking — The Upside of GMOs (Genetically Modified Organisms)

Get the only spam-free daily newsletter about wearables, running a "maker business", electronic tips and more! Subscribe at AdafruitDaily.com !



22 Comments

  1. Awesome!!
    Now only if we had a way to record our movements with the Kinect in order animate a 3D computer model. Perhaps for quick and easy development of NPCs (Non-player characters) in video games. *hint*

  2. Hey, I should actually thank you guys. Since I don’t own an Xbox, I couldn’t have done it without the USB protocol dumps you extracted. So, many thanks!

  3. Anyone else thinking that all that’s needed now is several Kinect working together to get a full 3D real time capture?

  4. Now you need multiple Kinectes to make a even better 3d model. .

    You could also use this to scan in an object to print using a reprap or Makerbot

  5. wow!

  6. cool so you can make 3D models

  7. So how about using 4 kinects together to get a full 3d image?!? Then, in theory, we could create full holograms!!! Heh, a unit built for gaming could lead to us being able to make holographic phone calls!!!

  8. Hooo boy that is completely awesome. This is really, really big for the 3D printing community man, you could use this to generate a really high-definition 3D scan of an object for recreation on a Makerbot! Or for game avatars and such, of course.
    Really, that’s excellent work. Thanks for sharing this, maybe I’ll actually get a Kinect without buying an Xbox just to experiment with this stuff. πŸ˜€

  9. Two Kinects for full 3-D!

  10. W0W just wow, the possibilities for this thing are endless now (not that they weren’t in the first place)

  11. W0W just wow, the possibilities for this thing are endless now (not that they weren’t in the first place)

  12. This actually made my jaw drop. Thank you Adafruit for setting up the prize. Everyone got so excited for the wiimote, now the Kinect is going to take off.

  13. It could be used as a (coarse) 3D-Scanner! There are a number of nice papers on combining multiple depth images like this to create a non-shadowed version of for example the box he held up (i.e. hold it to the camera, rotate so it sees every face, combine the images: get a textured 3D-model of the box.

    It even works with only 2D-webcam images (by building a 3D-model as the object is rotated, once it’s been seen from all sides you can also calculate its depth), but should be considerably easier when you already have the depth channel…

  14. Next use 2 or more Kinects, and combine the data, to eliminate shadows/holes. πŸ™‚

  15. I wonder how difficult it would be to use and synchronise multiple Kinect devices to capture a full 3D textured image of a space. The hardest part would be the image registration (matching up the physical position of the objects), but the depth system would make that a lot easier.

    Also, I wonder if multiple Kinect units can operate in the same space, or whether the IR grids they project would interfere with one another.

    Anyone else thought about this?

  16. multiple connects. four walls, four connects.

  17. That’s truely cool!

  18. So how about adding another Kinect on the other side of the person and show both sides of the person as you rotate around? This is some really cool stuff. Keep up the great work.

  19. Some one should make a 3d projection modeling app. The kinect would be the perfect utility to make this happen.

  20. If you use more than one Kinect, I don’t think they’ll be able to distinguish which Ir lights are from which one. You would need to hack the hardware to maybe flicker them so that only one is on at a time.

  21. This is excellent. I’m already playing with the thing more than perhaps I should, to then turn it into a 3D capture device for 3D modelling would be great, as a good part of my time will be spent building 3D models for my final year project πŸ™‚

  22. Hey great work! I have been thinking a lot about AI recently and how our brains work — it would be neat to have a 3D capture system like yours working with an algorithm that could do a "fuzzy" guess at what the missing/black geometry and textures should look like (perhaps with bayesian networks/probabilities & thousands of real life samples). This would enhance the effect tremendously. As a 3D designer, I could make some pretty good guesses at what the missing geometry would look like, so why can’t a computer? I know this is a HUGE task, but I wanted to throw it out there. πŸ™‚

Sorry, the comment form is closed at this time.