There are libraries for recognizing individual video frames, but I need to recognize an object in motion. I can recognize a person in every single frame, but I need to know if the person is running or waving. I can recognize a tree in every single frame, but I need to find out if the tree is swaying in the wind. I can recognize a wind turbine in every frame, but I need to know if it's spinning right now.
So the question is: do technologies, libraries, concepts, or algorithms exist for recognizing objects over a certain period of time? For example, I have a series of several pieces of frames, each of which exactly has a person, and need to find out if person are walking or waving their hands.