» CamShiftTracker (FreeFrame DShow9)
This site relies heavily on Javascript. You should enable it if you want the full experience. Learn more.

CamShiftTracker (FreeFrame DShow9)

Info This node comes with the addon pack. Get it here.

Here you are supposed to see an image (rendered with SVG) describing the pins of this node. However, this does not work with Internet Explorer. You may want to consider using another browser, like FireFox, Chrome, Safari or Opera that do support SVG.


related discussions

start a forum thread about this node

about CamShiftTracker (FreeFrame DShow9)

open source


When you use CamShiftTracker the most important step is the fine tuning of the tracking parameters.

These are:

  • Size and orientation of initial ROI
  • Saturation min, Value min and max

To achieve an optimal tracking output, these parameters have to be adjusted to the lighting situation of the video scene (or vice versa).

If possible, turn off automatic gain control (AGC), automatic brighness ajustment etc. on your camera.

During fine tuning of the parameters you should have Show Backproject set to 1, so you can see which pixels of the image are considered as belonging to the object (i.e. those which are white). Also, set Show ROI to 1, to see the search frame.
When nearly all pixels in the area of the object are white and the rest of the image is black, you're done.

Like a good linear minimizer, adjust one parameter at a time, then continue with the next one. Playing around will give you a feeling for this...

Be sure that your object lies in or, ideally, fills the initial ROI when you initialise the tracker by banging Init Tracker. You will get the best results by using a white background. Since white (background) pixels have a very low saturation, you can kick 'em out by increasing the 'Saturation min' pin value.
Adjust Value min and Value max to the range of your object brightness:

dark very low V1 (near 0.0) low V2
bright high very high V2 (near 1.0)
different shades low high

After each tuning step, init Tracker again to submit the tracking values and refine the object search.

If all that doesn't help, you might have to improve the lighting situation.
Think about illuminating the object with a lamp, or adding a diffuser if the video image is too dark and the colors don't come out right.

anonymous user login


~24h ago

phl: Node based Spotify playlist generation anyone? http://smarterplaylists.playlistmachinery.com/

~1d ago

dominikKoller: Is there a reason for two Delaunay nodes?

~2d ago

joreg: @synth since a year beta34.2 what makes you unsure?

~2d ago

synth: What is the latest production ready/capable vvvv version? Thank you!

~3d ago

third paradise: Computers Are People, Too! 1982 Fairlight bit https://www.youtube.com/watch?v=BXbrzI-8wck

~3d ago

guest: someone in ADE?

~3d ago

sableRaph: Copycat challenge of the day. Don't look at the code! https://www.openprocessing.org/sketch/374249

~3d ago

mediadog: @mino same company has ToF depth camera with network connection: http://product.inrevium.com/tof.html - thanks for link!

~3d ago

mino: @tonfilm and 1000fps projector now on sale http://www.teldevice.co.jp/pro_info/2016/press_161019.html

~4d ago

tonfilm: Dynamic projection mapping onto deforming non-rigid surface https://www.youtube.com/watch?v=-bh1MHuA5jU