“iterative hypothesis testing”
Iterative Hypothesis Testing for Multi-object Tracking with Noisy/Missing Appearance Features
Abstract: This paper assumes prior detections of multiple targets at each time instant, and uses a graph-based approach to connect those de- tections across time, based on their position and appearance estimates.
Christophe De Vleeschouwer
11th Asian Conference on Computer Vision, Detection and Tracking in Challenging Environments Workshop, Daejeon, Korea, Lecture Notes in Computer Science, Vol. 7724-7727