If you don't have pure background then you will end up with more and false keypoints. MK-RoD is prone to false SIFT keypoint matches. So simply, the algorithm will not work if you use background.
Hi Caglar, you have done a great work, I have a question. Why is it recommendable to use a pure black background? I mean, what happens with the algorithm and the keypoints if you don´t have that background?
Hi all, i'd like to know if anybody has tried to use this algorithm on very large datasets (like point clouds with 300000-500000 points) and which options are recommended to minimize calculation times.
thank you all