[Bharath] recently uploaded the source code for an OpenCV based pattern recognition platform that can be used for Augmented Reality, or even robots. It was built with C++ and utilized the OpenCV library to translate marker notations within a single frame.
The program started out by focusing in on one object at a time. This method was chosen to eliminate the creation of additional arrays that contained information of all of the blobs inside the image; which could cause some problems.
Although this implementation did not track marker information through multiple frames, it did provide a nice foundation for integrating pattern recognition into computer systems. The tutorial was straightforward and easy to ready. The entire program and source code can be found on Github which comes with a ZERO license so that anyone can use it. A video of the program comes up after the break:
i used to use openspace 3d for making a ball maze by tilting a book with markers and it used opencv i think.
Ooh, better watch out, Bharath. When Jack Crossfire sees this he’s going to cry like the little bitch he is, and then strangle you with his ponytail.
What about ARuco ?
http://www.uco.es/investiga/grupos/ava/node/26
Seems more powerful and elaborated more to me.
Don’t see the point of using Bharath’s implemention right now.
Also ARMA, ARToolkit+, and ofcourse Qualcomm Vuforia :P
“and easy to ready” ?. Also, although this particular code uses ZERO licensing, for the whole system it’s probably not entirely correct, as it uses OpenCV, which is released under a BSD license, so at least that acknowledgement should be done when using the AR software.
Awesome stuff, but it reminds me that I saw this recently;
https://www.youtube.com/watch?v=2YnIMfw6bJY#t%3D17
Open source camera based 3d position tracking, but can work on both large and small scales. Incredible stuff.
Thanks for this link. I’ve been looking at marker-less camera trackers for a side project and this looks promising.
Moving the camera then gets you simple 3-D scanning of objects, right?
https://github.com/uzh-rpg/rpg_svo
nice, finally something to rival (and beat) PTAM
first impression difference I can see:
-they pick more robust features (sharp edges / corners)
-they keep good features
Wonder how it acts in scenes with moving objects.
If this is open sourced where is the source?
https://github.com/uzh-rpg/rpg_svo
Nice stuff ! Chech out my OpenCV #AR demos utilizing #Processing ! http://www.youtube.com/playlist?list=PLF9C3irJK3xh9U_PcK-X07zd-5F0gDsYw
Been looking for stuff like this for a LOOOOOONG time
Web page is typical geek garbage . Does not bother telling the uninitiated whats supposed to be going on in the video or the software . Utter crap
Ibrahim has a start on this for OpenCV Cam using a new pattern matcher. Sneak peek: https://www.youtube.com/watch?v=bDtCJQspanM
… alas, more work to be done… another algorithm needed it seems… but, Ibrahim’s working on it. :)
Does it works whit raspberry pi 2?