I was wanting to create a Haar trainer for detecting a pair of scissors. I know Haar training is good for detecting faces but is it just as good for objects like scissors or is there something else I should be using other than opencv for detecting every day objects?
I'm also currently working on a project to detect various kind of things. If you are willing to invest time in tweaking the training process and get a lot of positive and negative image, yes it's a good technique. Don't forget that it detect object in a "almost fix position". For example, the face detector don't detect face that are not align perfectly with the webcam. So if you want to detect scissors in various angle, you will need to create at least classifier (xml) for at least the angles that will mostly be visible from the camera.
I'm not aware of any "out of the box" detection technique for scissors but you can always use your imagination and use some picture manipulation to extract a contour and recognise de shape or maybe some background suppression if the camera is not moving.
Have fun!
Asked: 2013-01-28 22:38:03 -0600
Seen: 852 times
Last updated: Jan 29 '13
I don't know but I think that object's should have small inclass variation and be without a background.
I am getting big haar training XML files and they arent detecting scissors. Perhaps I need to do without a background like you say. I've done some googling for the term 'inclass variation' but havent found an explaination. Does it mean I shouldnt have too many different types of scissors?
No, I mean that scissors should "be similar" like faces for example.