Tracking Method
Object tracking, a core computer vision task, aims to identify and follow objects across video frames, addressing challenges like occlusion, rapid movement, and unknown object categories. Current research emphasizes robust feature learning, efficient memory management (e.g., sparse memory approaches), and the integration of diverse data sources (e.g., sensor fusion from drones). These advancements are improving tracking accuracy and speed across various applications, including autonomous systems, wildlife monitoring, and industrial automation, with a growing focus on handling complex scenarios and minimizing human intervention.
Papers
BootsTAP: Bootstrapped Training for Tracking-Any-Point
Carl Doersch, Pauline Luc, Yi Yang, Dilara Gokay, Skanda Koppula, Ankush Gupta, Joseph Heyward, Ignacio Rocco, Ross Goroshin, João Carreira, Andrew Zisserman
MobilityDL: A Review of Deep Learning From Trajectory Data
Anita Graser, Anahid Jalali, Jasmin Lampert, Axel Weißenfeld, Krzysztof Janowicz