mistracking

Mistracking refers to the failure of a tracking system to maintain correct alignment with a reference target, path, or dataset. It occurs when a tracker loses lock, drifts away from the intended target, or produces inaccurate measurements, resulting in unreliable or misleading output. The term is used across domains such as media technology, computer vision, navigation, and logistics to describe different manifestations of tracking error.

In media technology, mistracking describes misalignment between read/write mechanisms and the data tracks on a storage medium or film sequence. This can cause audible distortion, rolling pictures, or frame skipping, and is often the result of mechanical wear, improper calibration, or tape skew. In vision, robotics, and surveillance, mistracking occurs when an algorithm fails to consistently follow a target, due to occlusions, abrupt motion, appearance changes, or clutter, leading to drift or complete loss of the tracked object.

Common causes across domains include calibration errors, sensor noise, interference, environmental conditions, and rapid or unpredictable project trajectories. Data association errors in multi-target scenarios, model drift, and inadequate feature representation also contribute to mistracking. Mitigation strategies focus on improving reliability through calibration, sensor fusion, robust data association, and predictive estimation. Techniques such as Kalman filters, particle filters, and reinitialization after a tracking failure are used to reduce drift and recover from temporary losses of lock, while infrastructure improvements and quality control help prevent mistracking in physical systems.

See also: tracking error, drift, occlusion, sensor fusion.