Some questions about the experiments

[liu-zikang]

Hello,it's reallly a nice work.But I have some questions about the results of experiments,I really want to know how to get the test results of other papers compared，such as dcp-v2 and so on.I am looking forward to your reply,thank you.

[yewzijian]

I modified the other paper's code to train on the partial visibility scenarios, and performed inference on them. It shouldn't be too hard to do so, you mainly need to change their data loader. For DCP_v2, you can find my trained checkpoints at #21 .

[liu-zikang]

Thank you very much for your reply, but I still can't get the relevant indicators of dcp-v2 given in your paper by using the trained model provided by you. Would you mind explaining how to achieve it further？

[yewzijian]

Are you able to get similar results under the clean scenario? My clean scenario is the same as the unseen categories in the Deep Closest Point.

[liu-zikang]

Yes，and the trained model provided by you is really useful,but I have difficulty in computing the mean isotropic rotation and translation errors.I realy want to know how to compute the mean isotropic errors.

[yewzijian]

You can use my code to compute those measures, e.g. by saving the transforms into a numpy file and calling eval.py with --transform_file your_numpy_file.npy.

The code to compute the metrics can be found in the eval.py. You can copy those over to DCP's code if you wish.

Zi Jian

[liu-zikang]

Thank you！