Deblur-NeRF icon indicating copy to clipboard operation
Deblur-NeRF copied to clipboard

Published 20 hours ago • verified publisher icon limacv

PSNR is not the same

Open 520jz opened this issue 1 year ago • 4 comments

Hello, really thank you very much for checking out my question in your busy schedule. In the paper, the PSNR of the BlurFactor dataset is 25.60, but the PSNR I got when I trained with the BlurFactor dataset turned out to be 36, why is it lower in the paper? Is it a problem with my training?

520jz avatar Mar 15 '23 07:03 520jz

The config file I use is tx_blurfactory_full.txt

520jz avatar Mar 15 '23 07:03 520jz

The command is：python3 run_nerf.py --config configs/blurfactory/tx_blurfactory_full.txt

520jz avatar Mar 15 '23 07:03 520jz

You may want to check on this link (https://github.com/limacv/Deblur-NeRF/issues/6#issue-1301430857), which is caused by me mistakenly leaving one config on when doing the evaluation. But I'm not pretty sure if this is the actual reason for such a significant difference in your case. Another possible reason is the randomness. As I find in my experiment that using different random seeds may lead to different metric values.

limacv avatar Mar 16 '23 15:03 limacv

You may want to check on this link (#6 (comment)), which is caused by me mistakenly leaving one config on when doing the evaluation. But I'm not pretty sure if this is the actual reason for such a significant difference in your case. Another possible reason is the randomness. As I find in my experiment that using different random seeds may lead to different metric values.

Ok, really appreciate your answer.

520jz avatar Apr 06 '23 03:04 520jz