Sign Up to our social questions and Answers Engine to ask questions, answer people’s questions, and connect with other people.
Login to our social questions & Answers Engine to ask questions answer people’s questions & connect with other people.
Lost your password? Please enter your email address. You will receive a link and will create a new password via email.
Please briefly explain why you feel this question should be reported.
Please briefly explain why you feel this answer should be reported.
Please briefly explain why you feel this user should be reported.
I have a set of objects that don’t intuitively fit into a cv::Mat, and I want to cluster them. I have a distance function defined between any two objects, and I’m already including OpenCV into my project, so it seems convenient to use its implementation.
So, my question is, given a defined distance function, can I use OpenCV’s k-means clustering implementation when the objects aren’t intuitively cv::Mat compatible?
Unfortunately OpenCV’s KMeans is hardcoded to handle only floatig point, single precision values. Anything else must be converted, if possible, to be used inside KMeans.
It should not be something impossible to send a distance function to a generic KMeans algorithm, but the current implementation does not allow it. All it can handle are multi-dimensional floating point feature spaces, stored in
cv::Mat