xuxw98/ESAM

Train&Test on my own dataset

Opened this issue · 1 comments

Hi, thanks for your great work!
I'm wondering if I can test the model (for example trained on ScanNet) on my own data? How to figure it out?
I'm trying to test the model's generalizability.
I'm not very clear about it. Hope you give me some extra information.
Thanks!

Hi, thanks for your interest! For testing on custom data, you can try our visualization demo. You can organize your data to have the same format with ScanNet then run the script for prediction.