In Multi-Object Tracking, 2 main skills are important:
1 — Multi-Object Tracking: Associating obstacles from one frame to another.
The first skill needed is MOT (Multi-Object Tracking). It's where you learn how to tell which obstacle from frame t is which obstacle from frame t+1. It includes Graph Matching algorithms, Deep Learning, and many optimization techniques...
2 — Kalman Filters: Predicting where the next obstacles are going to be.
The second skill is Kalman Filters... in which you learn how to estimate velocities, how to predict the next positions, and how to estimate the motion of obstacles through time.
Ultimately, you want to be able to TRACK BOUNDING BOXES through time, build an OBJECT, and PREDICT their next positions. This is what we call Dual Tracking.
Course 1 — LEARN OBSTACLE TRACKING: The Master-Key to Become a Self-Driving Car Professional
Course 2 — LEARN KALMAN FILTERS: The Hidden Algorithm that Silently Powers the Future
BONUS — The LAB: Bounding Box Tracking with Kalman Filters
€375 or 3 monthly payments of €135
THE TRACKING PACK (Late 2020)
Closed
Bundle
3 Products
Fuse the better of two worlds to build robust tracking applications.