Dear organisers,
Thank you for your effort to organise CLIC again. I had a question regarding the perceptual video quality assessment track.
From the task description, I understand that we should develop No-Reference (NR) quality metrics, since the reference video won't be provided for the test set. However, I noticed that the reference videos are provided for the training data, and that VMAF, a Full-Reference (FR) metric appears on the leaderboard.
Therefore, just to be completely sure, I wanted to double check with you whether the task is NR VQA. Thank you very much for your help.
Best regards,
Danier