How much harder is it to do ML for video compared to text/images? [D]

Hey everyone -- I've been seeing a lot of companies recently pop up doing ML for video in many capacities. Some are generative video (like Stable Diffusion's new model or Pika) or some are doing AI video editing (like Veed or Opus Pro). I'm sure there are a million ML/video companies out there that I don't know about in fields like advertising, sports, entertainment etc.


What I'm curious about is how much harder is it to build models for video …

ai video ai video editing companies diffusion editing generative generative video hey images machinelearning pika stable diffusion text video