-
|
Thanks for your great work!
I wonder :
Are there any other steps I need to follow? |
Beta Was this translation helpful? Give feedback.
Replies: 8 comments
-
|
30.37 is the PSNR on the test set, which is not the same as the validation set. Only the organizer has the GT of the test set. This model is the model we used for the competition. |
Beta Was this translation helpful? Give feedback.
-
|
May I check for the test set, do I need to specify additional parameters in the testing pipeline for ensemble (×8) testing? (including Flip and Rotation). Thanks! |
Beta Was this translation helpful? Give feedback.
-
So far we did not implement it. You can modify the basicvsr.py for the ensemble. I will add support to it later. Thank you very much for letting us know your need :) |
Beta Was this translation helpful? Give feedback.
-
|
Thank you! |
Beta Was this translation helpful? Give feedback.
-
|
For the validation set, the PSNR for unprocessed video is 27.09, the 28.67(+1.58) from basicvsr++ is of the same improvement level as the test set. (no ensemble testing) In the nitire report, for the test set, the unprocessed video PSNR is 28.34 compared to 30.37(+2.03) of basicvsr++. (8x ensemble testing). Thanks for your help! |
Beta Was this translation helpful? Give feedback.
-
|
Hi @kai422 and @ckkelvinchan , Can I ask how to get access to the validation raw data? The NTIRE challenge only seems to be providing the compressed data for validation. Am I missing something here? Please suggest. |
Beta Was this translation helpful? Give feedback.
-
Please check this: https://github.com/RenYang-home/NTIRE21_VEnh |
Beta Was this translation helpful? Give feedback.
-
|
Thank you so much @kai422 ! |
Beta Was this translation helpful? Give feedback.

Please check this: https://github.com/RenYang-home/NTIRE21_VEnh