16-bit buffer SuperResolution Scaling

I am trying to change the sample code to work on 16 bit images, as it is currently set to 8bit. Is that a limitation of the super resolution feature, or can it work on a 16 or 32 bit float buffer as well?

Seems these forums are not really that active, so hoping someone else who has played with it knows a bit more.

The docs are non-existent, makes it hard to know what is available and what is not.

Clint Hanson

BUMP. Would really like to use this on a project in the next month.

The NGX SDK is 8bit. We have a newer SuperRes network as part of the Video Effects SDK https://developer.nvidia.com/video-effects-sdk-windows which uses float.



Thank you Andrew! Any idea how long the approval process should be expected to take? I applied a few days ago - no activity. Just curious (and impatient) :-) Thanks!