zju3dv / PGSR

code for "PGSR: Planar-based Gaussian Splatting for Efficient and High-Fidelity Surface Reconstruction"
https://zju3dv.github.io/pgsr/
Other
467 stars 26 forks source link

How to evaluate my custom dataset #14

Closed DYH200009 closed 2 months ago

DYH200009 commented 2 months ago

When I used my own dataset to execute the metrics.py file, I found that I could not get the corresponding metrics and my test folder was empty. How to perform the calculation correctly. This is my cmd : python metrics.py -m out_path

danpeng2 commented 2 months ago

Hello, thank you for your interest in our work. If the test folder is empty, you might need to check whether the --eval command was included during training.

DYH200009 commented 2 months ago

Thanks bro , work