search

LimHyungTae / patchwork

SOTA fast and robust ground segmentation using 3D point cloud (accepted in RA-L'21 w/ IROS'21)
MIT License
489 stars 75 forks source link

The different between code and paper #17

Closed HMX2013 closed 2 years ago

HMX2013 commented 2 years ago

Thank you for your excellent work. In your paper, If the probability of Ground Likelihood Estimation is larger than 0.5, than ˆGn belongs to the actual ground. However, I noticed you don't do this step, Besides, the calculation of equation （10）and （11）didn't happen. I want to know whether it's just some kind of simplification.

LimHyungTae commented 2 years ago

I confess that the equations are quite redundant haha... But the equation exactly says the same things.

For more details, please refer to wiki: https://github.com/LimHyungTae/patchwork/wiki/3.-Methodology-of-Patchwork

LimHyungTae commented 2 years ago

P.S. (-) symbol in (10) is wrong in https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9466396.

It is modified in arxiv version as follows:

image

It was my terrible mistake...I'll try my best to submit correction file ASAP.