[mockery]

This is cool, but the frame-to-frame variance is distracting. I really want to see this reimplemented with temporal constraints a-la this paper:

https://www.youtube.com/watch?v=Khuj4ASldmU

[pvinis]

I hadn't seen that before. This looks great.

[thope]

I wish I had time to set up a computer and compile and run this

https://github.com/manuelruder/artistic-videos

[grabcocque]

I want to see them all now.

[onetwotree]

Oh wow, that looks amazing! I wonder how long it takes to transform those clips though...basic neural style transfer takes a few minutes on a standard gaming rig, so it probably takes a big machine or a few hours to do this.

[robbiep]

Wow