Orientation (yaw) estimation of AR drone using OpenCV
Hi,
I am working on a project that involves using stereo cameras for obtaining the 3D coordinate of a drone. I am currently trying to obtain orientation (yaw only) estimate of the drone using a pair of external (manually) calibrated stereo cameras.
My Question:
Are there any ways to do this in OpenCV ? I have seen this tutorial but it does not work for 3D non-planar objects.
Note: Kindly note that while there are some existing posts (like this ) but they are mostly on triangulation for the 3D coordinate, while my question here is on finding the orientation (yaw only) using external stereo cameras.
Thanks!
Isn't your yaw angle always relative against the current position of your drone, so it will change constantly? How will you cope for this with your static stereo setup? Your initial goal simply seems unachievable.
I'm not sure why you say it sounds infeasible, but motion capture cameras (see this ) have already been made.
yes but they capture motion right? They do not estimate the yaw angle, they estimate the relative movement against the camera setup, while from your question, I got the impression you want the yaw angle of the drone itself, which is an extra step... could be wrong though.
And to add to that, your video uses IR trackers. they simply track a concise blob throughout a 3D space. That lower level then what you try to achieve.
I'm not sure what you are suggesting, but I believe that obtaining the 3D-2D point correspondences is the key in obtaining the 6D pose of an object, from which yaw angle can be extracted.