First all image corners are found. | |
Next these corners are tracked in 3D providing a sparse 3D data set. From this set a 3D surface can be created by triangulating between the tracked corners. | |
Finally, this 3D surface can be used in a variety of ways, for example, the creation of a contour map of the 3D world. |
Papers published on DROID include [Harris 87], [Harris 88] and [Charnley89].