Two frame input -> one frame output video interpolation


I’m completely new to pytorch and

I’d like to create and train network that takes in two frames from a video and outputs one new frame. What aspects of the library should I be using? Can I use the Datablock or databunch api for this? Any other suggestions?