Repository navigation
Why not using 512 x 384 images of koniq10k to avoid resizing? #22
Description
Activity
- changed the title
[-]which koniq-10k data do you use for training?[/-][+]Why not using 512 x 384 images to avoid resizing?[/+]on Dec 9, 2020 - changed the title
[-]Why not using 512 x 384 images to avoid resizing?[/-][+]Why not using 512 x 384 images of koniq10k to avoid resizing?[/+]on Dec 9, 2020 Sure you can use 512x384 images for directly training, it's just because we only downloaded the images of 1024x768 resolution...
Reacted by Shuyue Jia (Bruce)
Hi, thanks for your IQA work!
I noticed that you resize the input image of koniq-10k dataset.
Since there are two resolutions of data in koniq-10k dataset, if you use 512 x 384 as input size, why not using 512 x 384 image size provided by koniq-10k ?