 |
Author |
Topic: marker recognition (Read 1282 times) |
|
bjoern_kah
|
marker recognition
« on: Dec 12th, 2003, 5:09pm » |
|
hello
i am working on some augented reality scribbles in proce55ing. some of my video-experiments are working (tracking brightest point, tracking changes, tracking colors).
now i am trying to identify markers and track their xy-position. does anyone tried this? the "Edge Detection"-demontration does not help me. is there any simple unprecise and quick&dirty way to do this?
(ie http://www.ecs.soton.ac.uk/~pass99r/research/mt/mt.html)
greetings
bjoern
|
|
|
|
|
REAS
|
Re: marker recognition
« Reply #1 on: Dec 12th, 2003, 6:21pm » |
|
hello bjoern,
i'm very happy to hear you're working on this. it would be a great piece of code for everyone.
a common library for this is called the Artoolkit. here is a pointer:
http://mtd.fh-hagenberg.at/depot/graphics/artoolkit/
+ casey
|
|
|
|
|
michael05
|
Re: marker recognition
« Reply #3 on: Dec 17th, 2003, 11:01am » |
|
for my diploma i need markers too. so i am working on the same problem.
hough and fourier transforms seem too be the keys to realize marker recognition. i got some books from the library to understand these transform (why do we not have math-courses at the university)..
artoolkit works on my powerbook. but i dont know if the java-bindings could be modified to work outside windows.
it would be great if anyone could describe the transforms...
|
°°°°°°°°°°°°°°°°°°°°
http://www.m05.de
|
|
|
|
