Multi-appearance Segmentation and Extended 0-1 Program for Dense Small Object Tracking

TL;DR

This paper introduces a novel multi-appearance segmentation and a multi-hypothesis tracking approach for dense small object tracking, improving speed and accuracy in complex scenarios with occlusions.

Contribution

It proposes a new multi-appearance segmentation method and an extended 0-1 program, along with a one-to-many constraint, to enhance dense small object tracking performance.

Findings

Demonstrates improved tracking speed and accuracy.

Effectively handles occlusions with new constraints.

Outperforms existing methods in dense small object scenarios.

Abstract

Aiming to address the fast multi-object tracking for dense small object in the cluster background, we review track orientated multi-hypothesis tracking(TOMHT) with consideration of batch optimization. Employing autocorrelation based motion score test and staged hypotheses merging approach, we build our homologous hypothesis generation and management method. A new one-to-many constraint is proposed and applied to tackle the track exclusions during complex occlusions. Besides, to achieve better results, we develop a multi-appearance segmentation for detection, which exploits tree-like topological information and realizes one threshold for one object. Experimental results verify the strength of our methods, indicating speed and performance advantages of our tracker.