Beyond the Survey: A Systematic Empirical Study of Detection and Association in Visual MOT

超越综述：视觉多目标跟踪中检测与关联的系统性实证研究

Abstract: This paper presents a comprehensive experimental evaluation and detailed analysis of state-of-the-art multi-object tracking algorithms, with an emphasis on quantifying the individual contributions of detection and association components to overall tracking performance.

摘要： 本文对最先进的多目标跟踪（MOT）算法进行了全面的实验评估和详细分析，重点在于量化检测和关联组件对整体跟踪性能的各自贡献。

Unlike existing surveys that primarily offer theoretical categorizations or taxonomies of tracking methods, our work adopts a rigorous experimental perspective grounded in publicly available implementations, providing practical guidance for researchers and practitioners in method selection and system design.

与现有的主要提供理论分类或跟踪方法分类的综述不同，我们的工作采用了一种基于公开实现方案的严谨实验视角，为研究人员和从业者在方法选择和系统设计方面提供了实践指导。

We introduce a unified pipeline diagram that consolidates the core components across the two main branches of visual multi-object tracking: tracking-by-detection and end-to-end deep learning paradigms, and systematically analyze the object detection, feature extraction, and data association modules.

我们引入了一个统一的流程图，整合了视觉多目标跟踪两大分支的核心组件：检测跟踪（tracking-by-detection）和端到端深度学习范式，并系统地分析了目标检测、特征提取和数据关联模块。

Through extensive empirical studies on standard benchmarks, including MOT16, MOT17, MOT20, SportsMOT, DanceTrack, and CrowdTrack datasets, we reveal critical insights: (1) detection quality dominates association strategy performance, with detector improvements yielding more than 10% gains compared to less than 5% from refined association strategies; (2) modern deep learning detectors paired with specialized re-identification models significantly outperform joint detection and embedding approaches; and (3) transformer-based end-to-end methods exhibit greater robustness to detection quality variations but at a substantial computational cost.

通过在 MOT16、MOT17、MOT20、SportsMOT、DanceTrack 和 CrowdTrack 等标准数据集上进行的大量实证研究，我们揭示了关键见解：（1）检测质量对关联策略的性能起主导作用，检测器的改进可带来超过 10% 的增益，而优化关联策略带来的增益不足 5%；（2）现代深度学习检测器与专门的重识别（Re-ID）模型相结合，显著优于联合检测与嵌入（joint detection and embedding）方法；（3）基于 Transformer 的端到端方法对检测质量的变化表现出更强的鲁棒性，但付出了巨大的计算成本。

Our findings from extensive experiments provide key insights into component-level effects in MOT, particularly the dominant influence of detection quality relative to association, while offering practical insights for designing and optimizing MOT systems under varying performance and robustness requirements. Code and experimental setups are available at this http URL.

我们从大量实验中得出的结论为 MOT 中的组件级效应提供了关键见解，特别是检测质量相对于关联的主导影响，同时为在不同性能和鲁棒性要求下设计和优化 MOT 系统提供了实践参考。代码和实验设置可在该网址获取。