I'm currently working on an image recognition software. At the moment I'm working with 2D planes and I use "findHomography" with RANSAC to validate the initial feature matches.
Next I want to use 3D point clouds. As I understand from the online documentation of "findHomography" it is not applicable to 3D points. My question is, what is the best way to validate the feature matches of 3D point clouds? Can use "estimateAffine3D"? What do I have to consider, when I move from 2D to 3D images?