When I install the additional libraries for CV in the opencv_contrib, all other libraries get compiled but not tracking. Any ideas why ?? No .lib files are created only for tracking.
I tried to install the additional libraries for tracking from opencv_contrib folder using the method mentioned in https://www.youtube.com/watch?v=vp0Ab....
But only tracking modules are not working the rest are working. If i go to the .lib folder then no .lib is created for only tracking but to all others it is created.
So any suggestions will be highly appreciated.
Do you by chance have it unchecked in cmake? All the BUILD_opencv_* boxes, you might have accidentally clicked it.
tracking depends on datasets, and datasets depends on face, so please enable those, too.
Thanks for the reply !!! I have enabled them. But when I click on configure in cmake the following two lines are displayed: " 1. Module opencv_datasets disabled because opencv_text dependency can't be resolved! 2. Module opencv_tracking disabled because opencv_datasets dependency can't be resolved! "
Plz tell me how to resolve this. I need tracking very badly !!!
... yea, and who knows, what the text thing depends on...
if you want my 2ct:
change the CmakeLists.txt in the tracking module, and make the datasets dependancy OPTIONAL
Thanks berak !!! Now MultiTracker class works but ran into one more error !!!
I followed the example mentioned in link of Opencv 3.1 http://docs.opencv.org/3.1.0/d5/d07/t...
but the "trackers.add(img,found)" is not working. Then how to initialize the trackers for the first frame ??? found is the vector<rect> for people found using HOG SVM.
plz help
Thanks a lot!!! It worked
@ManuVISION, nice to hear that, but can you please start a new question for your new problem ;)
(else it gets pretty messy here)
@berak: are datasets and face modules dependencies really needed for the tracking module? I recently found myself with this same problem, and disabling the dependencies on the cmake config allowed me to correctly compile the tracking module without any issue...
@LorenaGdL,
yes, that's what i'd say, too. (it's only thosetwo samples, that rely on the datasets dependancy.
let's see, if making that dependancy optional works
(would spare some folks exactly that trouble in the future..)