search

zhangxy0517 / 3D-Registration-with-Maximal-Cliques

Source code of CVPR 2023 paper
MIT License
421 stars 45 forks source link

calculate the optimal 6-DoF rigid transformation #4

Open 2019EPWL opened 1 year ago

2019EPWL commented 1 year ago

Thank you for your excellent work! I have a question about the optimal 6-DoF rigid transformation. After obtaining a set of 6-DoF pose hypotheses, how to calculate the optimal 6-DoF rigid transformation, and the s(c_i) in equation(6) refer to which?

And in my opinion, the correspondence is only calculated by the initial feature match and not updated after, is the initial correspondence affecting the MAC method performance?

zhangxy0517 commented 1 year ago

For the first question: After obtaining the hypotheses set, we use the metric (MAE for example) to calculate an evaluation score for each hypothesis, then the one with the largest score will be the optimal rigid transformation. s(c_i) refers to the metric score of correspondence that constitutes the evaluation score. You can learn the detailed process from the function evaluation_trans in PCR.cpp.

For the second question: Yes. The quality of the input affects the output result. The higher the initial matching inlier ratio, the higher the probability of successful registration, and the smaller the error.”