I like toe detact my horse in het stable by using a raspberry pi with cam to detact what she is doing in her stable. i like OpenCV to detact if she is :
eating hay of bix standing laying down laying flat (with her head down also) Drinking stable lot occupied
I understand i have to make an XML file for each situation with 5000 images? I like to do this by videofile. I would like to know how to make an XML file by using de frames of a video.
ps i am new at Linux an opencv