Greetings NGX team! There was mention of the ability to supply your own training dataset in Thomas True’s 2018 Siggraph talk. Is this something I can do now or will it be exposed in the next SDK update?
I’m working with video that is fairly tricky so I’m not seeing much of an improvement using the current dataset however I do have a significant amount of high resolution training material for this footage ready to be utilised :)
Cheers in advance!