Z-ENG: Marker based positioning using image processing
The task is to develop an algorithm that is able to detect polygon-like shapes in camera images. These detected polygons (quads) and the reference ones are then used to estimate the 3D viewpoint (position, orientation) of the camera.
Several approaches can be applied to solve the problem (line fitting, thinning, corner detection, Hough-transform, contours etc). Also several methods can be selected and compared. For viewpoint estimation the Robust planar pose algorithm is suggested.
Platform: C# + OpenCV (opencvsharp)