» Manipulating Faces
This site relies heavily on Javascript. You should enable it if you want the full experience. Learn more.

Manipulating Faces

demo face tracking kinect dx11
Credits: woei, joreg

about

Besides its full body/skeleton tracking capabilities the Kinect also features a dedicated face tracking functionality. With it you can easily identify individual parts (eyes, nose, mouth) of a face and then manipulate, replace, extend those using texture- or geometry-based techniques.

The download comes with 13 different demos on what you can do with the tracking data.

Note:

  • Requires b32.1 + addonpack + dx11 pack
  • Those demos will not work with Kinect2 out of the box but should be fairly simple to adapt once the Kinect2 nodes also features face-tracking capabilities.
  • download

    ManipulatingFaces.zip
    28.07.14 [14:46 UTC] by joreg | 1579 downloads

mino 29/07/2014 - 07:39

thanks for sharing! but FullscreenQuad and some modules are missing here.

esnho 29/07/2014 - 15:29

Seems good but no Kinect no ffffun... :(

joreg 29/07/2014 - 15:53

@mino: i added requirements above. does that help?

@esnho: in most samples you can easily use FaceData (Kinect Microsoft Fallback) to simulate the face.

mino 29/07/2014 - 16:33

it worked! i like Celeb and UVReplace one ;)

mpenna 20/08/2014 - 16:57

Nice demos indeed... thanks for sharing! I have some experience programming for the Kinect but on other platforms - vvvv is quite new to me. The "makeup" patch is one that really caught my eyes. I believe it's possible to try to use another face shot as reference image, right? But how to go about editing the face markers? The patch loads them from the "face.vpe" file, right? Do we have to edit this file by hand or is there some tool that can save these points to a file?

joreg 20/08/2014 - 17:15

the patch is using a PointEditor (3d Persistent) internally that allows you to manipulate the points directly. have a look at the helppatch of that node to learn how to operate it.

mpenna 20/08/2014 - 23:50

@joreg: I'll take a look at it... thanks!

io 04/09/2014 - 23:18

Not sure why but my face is not being recognized, all nodes are fine, Kinect is working, maybe I am just too ugly?
Simone

io 05/09/2014 - 22:23

SOLVED: you have to be in full sight (a least torso) of the camera, just the head isn t enough...

anonymous user login

Shoutbox

~2d ago

joreg: @eglod vvvv is international! wenn du hilfe auf deutsch brauchst, dann komm am besten in unseren deutschen chat: https://riot.im/app/#/room/#vvvv-at:matrix.org

~2d ago

eglod: In allen möglichen Sprachen gibt es VVV. V4 ist deutsch, warum nicht in deutscher Sprache. eglod

~2d ago

eglod: In allen möglichen Sprachen gibt es VVV. V4 ist deutsch, warum nicht in deutscher Sprache. eglod

~3d ago

joreg: Reminder: First #vvvv meetup in #hamburg this friday: vvvv-hamburg-meetup-1 Hope to see some of you there!

~5d ago

AKa-visuals: Hi guys. Everyone guess when xenko engine would be avaliable to play around in vl ... Isnt?

~13d ago

joreg: and right after the intro join us for our second Patching Circle: patching-circle bring your projects and problems and patch along

~13d ago

joreg: Reminder: This thursday: Free #vvvv gamma intro course: free-vvvv-intro-workshops-this-summer-in-berlin