Qifeng Chen
Qifeng Chen
The evaluation is essentially user study on Amazon Turk. I will update the evaluation code soon. It just creates javascript for Amazon Turk. There is no automatic measurement.
Hi, I just upload the Amazon Turk scripts on Github. See folder "mturk_scripts".
@tesfaldet I forgot to put the Javascript code on GitHub. I just upload example code in "mturk_scripts".
I think you can use Rawpy as one of the baselines as the traditional pipeline. Different cameras have their own ISPs.
Thanks!
Oh, then I think we should fix the misalignment problems on some images in the dataset. @cchen156
Please train on patches of sizes such as 512x512.