This forum is disabled, please visit https://forum.opencv.org
 | 1 | initial version |
A Kalman filter does not know of your points's x and y coordinates. It can handle any N-dimensional space, but it's up to you to define it. In your case, you have 4 points with 2 coordinates, that means 8 different inputs to the filter. So you define a 8x1 input matrix, and assign the 1st coordinate to p[0].x, 2nd coordinate to p[0].y, and so on.
