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
filter3x3 already supports this, so we'd just need to decide how to combine the per-channel magnitudes. It's not obvious what the correct approach is here.
The text was updated successfully, but these errors were encountered:
I think we shouldn't bother trying to do anything clever. Instead, we should just provide a method which computes per-channel gradients and applies a user-provided function to combine them, with two standard options provided:
CombineChannelGradients {
Max, // Use max of per-channel gradients
Mean, // Use mean of per-channel gradients
}
From some quick googling these both appear to be reasonably common.
See https://mht.technology/post/content-aware-resize/
filter3x3 already supports this, so we'd just need to decide how to combine the per-channel magnitudes. It's not obvious what the correct approach is here.
The text was updated successfully, but these errors were encountered: