[D] Transform (x,y) coordinates to 1 single feature

Hi

I'm very new to Machinelearning, and I want to use an SVM to classify a pose. I have a list of features (keypoint of the pose) in x and y coordinates. Since they are consistent, I want to use them as a single feature. I was wondering what the best method is to change the (x,y) coordinates to 1 feature.

Thanks in advance

machinelearning