no code implementations • 10 Mar 2020 • Alexandre Pineault, Guillaume-Alexandre Bilodeau, Gilles Pesant
Because our focus is on the data association problem, our MOT method only uses simple image features, which are the center position and color of detections for each frame.