Robots learning to cook by watching YouTube videos (PDF)

In order to advance action generation and creation in robots beyond simple learned schemas we need computational tools that allow us to automatically interpret and represent human actions.

This paper presents a system that learns manipulation action plans by processing unconstrained videos from the World Wide Web. Its goal is to robustly generate the sequence of atomic actions of seen longer actions in video in order to acquire knowledge for robots.

Well, not cook completely per se, but able to watch videos and identify grasping techniques and objects and tools used and replicating that is quite a start.