Event-based data allows efficient image segregation through event-based feature and velocity detection using neuromorphic hardware.
The video above shows image seggregation based on the fact that objects closer to you appear to move faster than objects further away when you are moving yourself, such as in a car or train. The set-up for this recording is shown below.
An artificial hedge is blocking the view of the camera that translates past it as shown in the raw data in the top left. By tuning event aggregation to different velocities, we can look at the foreground (top right), background (bottom right), or at the middle distance (bottom left).