How you retain NR SOTAs?

guanghaoyin / CVRKD-IQA

Pytorch code for our AAAI2022 Oral paper "Content-Variant Reference Image Quality Assessment via Knowledge Distillation"

MIT License

56 stars 8 forks source link

How you retain NR SOTAs? #3

Open uncontinuous opened 1 year ago

uncontinuous commented 1 year ago

"For fair comparisons between FR/NR/NAR methods, we retrain those NR/NAR-IQA SOTAs with the same experimental setting as ours. "why are there NR methods needed to be retrained? and how they retained with the same experimental setting as yours? Thanks!