--=-ao/BkP05wDPSmvgdWrRf
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Hey Daniel and others,
Im currently doing measurements on jitter, accuracy and other properties
of marker tracking in ARTK+. However, during some of my tests i
discovered depth measurements that didnt make sense to me.
My current Jitter test takes 100 frames of a stationary markerboard
about 50 cm from a 640x480 dragonfly express camera. The resulting depth
measurment (z-coordinate of the marker) of some markers vary widely
between frames, over a range of 40 mm. This would mean that the
difference in size of the marker between the minimum and maximum frame
is quite large (around 8%). However, if you take a look at the markers
in question, the difference is minimal (a bit of grey value on the
borders), as you can judge from the images in the attached archive.
My question is if i have missed some step in ARTK+ that distorts the
image besides the lens distortion which could explain the behavior, or
if it is the result of something in the pose estimation algorithm?
In the archive i have included my current camera calibration file
stripped of everything but the resolution and focal length. The main.cpp
file is the simple example program included in the ARTK+2.1.1
distribution, slightly modified to have a fixed threshold, apply no lens
distortion and use the original pose estimation algorithm. To compile
it, simply put it in the same spot as the original.
The images itself have been whitened out to show only the marker in
question. If you run the simple example program on them, they will
(should) return z values of 512.97 and 473.34. If you look at them with
an editor, you can see that they are almost identical in position and
size.
I hope someone can help me with this,
regards,
Tijs de Kler
University of Amsterdam
--=-ao/BkP05wDPSmvgdWrRf
Content-Disposition: attachment; filename=test.tar.gz
Content-Type: application/x-compressed-tar; name=test.tar.gz
Content-Transfer-Encoding: base64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--=-ao/BkP05wDPSmvgdWrRf--
|
