Diferent objects detection in one picture

asked 2019-10-30 10:43:05 -0500

Bilityuk gravatar image

updated 2019-10-30 10:47:06 -0500

Hi, i am trying to find the right way of how to detect different objects in one picture with several classifiers simultaneosly. I know how to detect face on the picture with "lbpcascade_frontalface.xml" default opencv classifier, but if i want to simultaneosly detect also "dogs.xml", "cats.xml" and so on... what is the best right way to do it? I am using C++ (opencv v 2.4.1). thank you for replies...

image description

edit retag flag offensive close merge delete



The picture your are showing is taken from the yolo object detector. Theres a opencv tutorial how to do this: https://docs.opencv.org/master/da/d9d...:

The model it uses was trained on the a coco dataset and has 80 classes: You can also do the thing without opencv by using python or c++: https://github.com/pjreddie/darknet/t...

holger gravatar imageholger ( 2019-10-31 06:01:50 -0500 )edit

To clarifiy things: It was done by using a CNN. The xml files you are talking about are are for the cascade classiifers(lbp and haar). This is a different technology used for the face detector examples. While this works fine it has its limitations. For example you cannot detect multiple objects at the same time (it only does binary classification)

holger gravatar imageholger ( 2019-10-31 06:15:53 -0500 )edit