Empace Wang
Empace Wang
Hi, I got similar results to you. I guess one reason may be the network haven't converged well. We may try more training epochs. Btw, have you solved the problem?
> With the current code in Line 84, it is not in compliance with Line 83. Should the correct one be as follows? > Line 84: self.cur_R = self.cur_R.dot(R) I...
@CaptainEven Thank you, and sorry for the delay. It is mentioned in the official document of KITTI dataset http://www.cvlibs.net/publications/Geiger2013IJRR.pdf, and it is 1.65m in that document. But we do not...
@JEFFCHANG0501 Hi, thanks, and sorry for the late reply. The reason for your problem may be the difference in the OpenCV version. I tried OpenCV '4.5.5', where the following code...
Hi, thanks and I am sorry for the late reply. I think you are trying to find an online solution, so you may need a C++ version scale recovery method,...