You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I noticed that there are some deformations in the synthetic image pairs (384×384) provided by your project page. Maybe you did some operations like resize? Do they have any impacts on your results?
By the way, in your paper, you said that these raw images are resized to 400×600. Why are they 384×384 currently?
The text was updated successfully, but these errors were encountered:
Yeah, I had the same doubt regarding the image size of synthetic images not being 400600. When I'm training on my own dataset, do I have to resize them to 400400 or can I use them in their original size 512*384
I noticed that there are some deformations in the synthetic image pairs (384×384) provided by your project page. Maybe you did some operations like resize? Do they have any impacts on your results?
By the way, in your paper, you said that these raw images are resized to 400×600. Why are they 384×384 currently?
The text was updated successfully, but these errors were encountered: