Reputation: 23
I have a project using 3d facial feature points from kinect sensor. I need to calculate distance between some points so that I get a distance that is invariant to scale, translation, rotation. I read that I needed to compute normalized distance between points but I don't know how to do that. I looked on google but didn't get a mathematical formula. If someone could help me or at least give me tips to documentation to do that. Thank you.
Upvotes: 0
Views: 3053
Reputation: 29724
You can calculate a normalized Euclidean distance between vectors x, y (from same distribution) with:
where s is a standard deviation of x and y. In general you can look at Mahalanobis distance
Upvotes: 3