The computer was taught to change the speed of only one object in the video

A team of specialists from Google and the University of Oxford has created a neural network that can change speed.

only one object in the video. She can distinguish between people and objects in video footage.

The algorithm can slow down or speed up movementsobjects independently of each other. At the same time, they continue to interact smoothly. To achieve this, experts divided each frame in the video into separate layers and taught machine learning algorithms to recognize people or objects in them.

The above example shows how the neural network synchronizes the jumps of individual boys in the video: