search

yulunzhang / RDN

Torch code for our CVPR 2018 paper "Residual Dense Network for Image Super-Resolution" (Spotlight)
546 stars 109 forks source link

Testing on "wild SR" = track 4 NTIRE2018 #10

Open sefibk opened 5 years ago

sefibk commented 5 years ago

Hi, I am interested in the NTIRE2018 track4 challenge (wild SR) This may be a basic question but to which of your degradation methods does this challenge refer? (BI\BD\etc.) To the best of your knowledge\assumption - would your network perform good SR on any random image (e.g. taken from a phone) or are there degradation limitations?

Thank you

yulunzhang commented 5 years ago

Hi, I think the degradation model in NTIRE2018 track4 (wild SR) is different from the degradation models used in the paper.

Currently, the method is trained in a super-vised way. It means we have to provide LR and the target output (namely Ground Truth) pairs for training. If the data are available, I think our method could learn good models regardless of the types of degradation. Thanks.