Multi-Object Tracking in Dynamic Scenes by Integrating Statistical and Cognitive Approaches
In this paper, the authors have addressed a quite researched problem in vision for tracking objects in realistic scenarios containing complex situations. The framework comprises of four phases: object detection and feature extraction, tracking event detection, integrated statistical and cognitive modules, and object tracker. The objects are detected using fused background subtraction approach along with feature computation. Next, the tracking events are inferred by finding spatial occupancy of moving objects. Third module is the key to proposed approach and the motivation is to tackle the tracking problem by axiomatizing and reasoning human-tracking abilities with associated weights.