search

IanYeung / RealVSR

Dataset and Code for ICCV 2021 paper "Real-world Video Super-resolution: A Benchmark Dataset and A Decomposition based Learning Scheme"
Apache License 2.0
115 stars 7 forks source link

about test result #4

Open sunyclj opened 2 years ago

sunyclj commented 2 years ago

excuse me，input size is 512*1024，after Video Super-resolution，the size is thr same as the input. i think the size should be bigger, x2 or x4 and so on.what is the reason?

sunyclj commented 2 years ago

When I entered my consecutive 50 frames , I got 50 frames of images, but the results were terrible ,the output is more ambiguous than the input.why?

IanYeung commented 2 years ago

When I entered my consecutive 50 frames , I got 50 frames of images, but the results were terrible ,the output is more ambiguous than the input.why?

Hi, for the resolution problem, please refers to other several issues. May I know what is your input?

fehimebetul commented 2 years ago

I agree with the @sunyclj, becuase SR is used for upsampling so, if we use LR and HR with the same size, What does SR algorithm do?