Hello All;
SmartSurface released SmartSurface CCV tracker engine based on NUIGroup’s CCV. This version supports 4 cameras, cropping of inline borders for alignment and mouse clicks.
It is released under GPL so you will get the source code as well (soon).
http://smart-surface.com/forum/viewtopic.php?f=8&t=3&p=3
more info u can get from getsmartsurface.com
we would like to thank NUIGroup for their great efforts in supporting multi-touch
* Your modifications allow for usage of one OR four cameras, not two, not three?
* Does the four-cam mode work by customizing the driver like AlexP does it, or by really stitching the images several (potentially different) cameras produce together?
* What operating systems are supported for that multi-camera mode?
* What version of CCV is this based on?
@dennda
* only one or four
* it is not like AlexP one, u can use any camera even u can use different types of cameras, so it’s not done on the driver level
* now windows only
* CCV1.2
If you make it so that it can also use 2 or 3 cams and supports OSX (and while you’re at it, also Linux ) I will happily use it instead of the stock CCV. (Also they should really consider merging it then.)
It’s very good that you didn’t do it on the driver level though.
If you make it so that it can also use 2 or 3 cams and supports OSX (and while you’re at it, also Linux ) I will happily use it instead of the stock CCV. (Also they should really consider merging it then.)
It’s very good that you didn’t do it on the driver level though.
@dennda, we plan to enable 2 & 3 cam in the near future and Support for Linux. Not too sure about OSX. But the source will be released soon, so u can take a look at it and suggest modifications. and my experience with OSX programming is zero..!! so any help theer would be appreciated.
(1) Can I used USB hub for 4 cameras connections? if so, any data bottom neck traffic issues? can I used mix blands of cameras?
(2) and any guildline or limitation of the use camera,
(3) I also try to register, the vertification ion is hard to read? failed few times, any way make it easier like by added numbers or some kind of equation instead?
If this is being built using CCV, openCV and openframeworks then programming it for one system should generally work on all 3 since the original CCV is cross platform.
instead of cropping off the edges of camera capture, would it be possible to make a program in which you can just set up the cameras like one would normally do, and then enter a calibration mode such as the one in CCV with calibration points? the program could detect overlapping areas, as well as stitch edges together in a neat seam. It would also automatically rotate and stretch or skew images so they fit together seamlessly.
I also have a question: how does Alex P’s PS3 multicam driver work? does it work with cropping, or like the method i just described, or some other method i am not aware of? I’ve never tried it because i don’t have multiple cameras. just curious, though i don’t plan on using multiple cameras soon.
@uncledj: try and let us know any way it was working with us for 32 frame/sec, plug 4 cameras i use uni-brain camera then u should be able to see the 4 cam, crop what u do not like and everything should be fine, regarding the registration there is many people register so far so give it second try
@Seth (cerupcat) : yes should work on any platform but so far we compile it by MS visual studio .net
@cclementi6: we will take that in consideration, about AlexP i have no idea i did not try it.
1) If you use a hub, in some cases, it has issues detecting the 4th camera. I suggest that you use the native built in USB ports in your PC. Till now, i haven’s detected any bottleneck. It was giving 32fps with 4 FireWire cams. and YES, you can use any blend of camera. FireWire+USB+...+…
2) It will use any camera that windows registers as a camera
3) Try again please
) I will happily use it instead of the stock CCV. (Also they should really consider merging it then.)
