Pixel value (ni, nj) to real world coordinate (x, y) (Python)

asked 2016-02-25 04:59:58 -0600

marvineffing gravatar image

Here is what I'm trying to do. I wish to translate a pixel from an image I captured from my IP camera to a real world coordinate on a field. I have calibrated the camera, and so I have all extrinsic and intrinsic variables. Secondly, I have defined the corners of the field so that:

  • the list clicked_corners_ij contain the pixel coordinates where the user has clicked defining the corners of said field, example: [[928, 284], [988, 1462], [120, 1426], [154, 334]]

  • the list corners_xy contain the real-world coordinates respectively: [[0, 0], [w, 0], [w, h], [0, h]] where w and h are the assigned width and height of the field.

So my question is what else do I need, and how do I use my intrinsic, extrinsic camera variables together with the provided lists? Do I need to estimate the pose as well?

Kind regards,

edit retag flag offensive close merge delete
add a comment