hi all
My problem is how to classify two actions: running and walking using the SVM method.
First I have 02 videos (the first action is running, the 2nd one is walking), after, I've transformed them into a sequence of binary images MC and MM (I have used the Contour detection ie: the 02 matrices contain only 1 and 0).

MC: is a matrix of dimension (120 * 160 * 65), 65 is "number of frames of the first video.
MM: is a matrix of dimension (120 * 160 * 87), 87 is "number of frames of the 2nd video.

issues:
1-how can i use SVM?how can form my data matrix and my groups vector?
2-should i use directly the matlab commands svmtrain, svmclassify and adapt them to my problem or I have to write my own code?
3 -is svmsacling important in my case?

thank you in advance