LITE让多目标追踪更快更省，实时融合外观特征
LITE: A Paradigm Shift in Multi-Object Tracking with Efficient ReID Feature Integration
- 将外观特征提取嵌入跟踪流程，无需额外模型与预处理
- 在MOT17上达43.03% HOTA，速度是DeepSORT的2倍
- 适合追求高效部署的工业级多目标追踪场景
提出轻量级集成追踪-特征提取（LITE）范式，通过将外观特征提取直接整合到基于检测的追踪流程中，消除推理、预处理、后处理及ReID模型训练开销。采用YOLOv8m等标准CNN检测器，实现无延迟外观特征提取。在经典DeepSORT基础上实现的最简LITE版本，在MOT17上达到43.03% HOTA，运行速度为28.3 FPS，相较DeepSORT在MOT17上提速两倍，在更密集的MOT20上提速四倍，且保持相近精度。新评估框架显示，传统追踪器在公平条件下仍具竞争力。代码将于发表后公开于https://github.com/Jumabek/LITE。
原文摘要 · Abstract (English)
The Lightweight Integrated Tracking-Feature Extraction (LITE) paradigm is introduced as a novel multi-object tracking (MOT) approach. It enhances ReID-based trackers by eliminating inference, pre-processing, post-processing, and ReID model training costs. LITE uses real-time appearance features without compromising speed. By integrating appearance feature extraction directly into the tracking pipeline using standard CNN-based detectors such as YOLOv8m, LITE demonstrates significant performance improvements. The simplest implementation of LITE on top of classic DeepSORT achieves a HOTA score of 43.03% at 28.3 FPS on the MOT17 benchmark, making it twice as fast as DeepSORT on MOT17 and four times faster on the more crowded MOT20 dataset, while maintaining similar accuracy. Additionally, a new evaluation framework for tracking-by-detection approaches reveals that conventional trackers like DeepSORT remain competitive with modern state-of-the-art trackers when evaluated under fair conditions. The code will be available post-publication at https://github.com/Jumabek/LITE.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。