We've developed an algorithm to anticipate visual events that may happen in the future

Anticipating outcomes is the root of intelligence. In computer vision, we investigate Predictive Vision with the goal to develop robust methods that anticipate the next events that may happen. Importantly, we focus on efficiently scaling learning algorithms to learn an extensive set of rules that enable intelligent vision. To solve this challenge, we develop predictive vision algorithms that learn these rules directly from large amounts of raw, unlabeled data. Capitalizing on millions of natural videos, we are teaching machines about the world by anticipating the visual future.

Research Areas