Tracking multiple objects fairly compared with same test settings
Tracking-by-detection in Multi-object Tracking: Survey and Experiments
Computer Vision and Pattern Recognition
Summary
Following inconsistent ways to judge how well multi-object tracking systems work, the authors review many methods that find and follow several objects in videos. They start with a simple tracker and test each addition carefully across different video sets to see what really helps. This lets them find the best balanced tracker and offers clear advice on building reliable tracking tools for real use. Their work helps separate useful improvements from just luck or mismatched comparisons.
What this means in practice
- •For surveillance system developers: Create better multi-object tracking systems by selecting proven, fairly tested methods to reliably track people and objects in video footage.
- •For autonomous vehicle engineers: Improve object tracking modules in self-driving cars by integrating robust tracking components evaluated across diverse scenarios.
A survey. It maps existing work.
Authors
Yujin Yang, Kyujin Shim, Kangwook Ko, Changick Kim
Abstract
Multi-object tracking (MOT) is an essential computer vision task that simultaneously tracks multiple objects in video sequences, with various applications in surveillance, autonomous navigation, and human-computer interaction. The tracking-by-detection (TBD) paradigm, which combines object detection with temporal association, has emerged as a leading approach, driven by innovative algorithms. Despite recent progress, fair evaluation of TBD-based methods remains a challenge. Many studies introduce modules such as similarity metrics, data association strategies, or motion models, but they are often evaluated under inconsistent protocols, with different baseline trackers, hyperparameters, and datasets. Such inconsistencies obscure the genuine contribution of each module and hinder objective comparison. This survey systematically reviews TBD-based MOT techniques, including similarity measurements, data association, camera motion compensation, and interpolation strategies. Starting from a minimal baseline tracker, we fairly evaluate the contributions of each method across diverse datasets and accumulate well-balanced methods. Our findings establish a strong baseline tracker and provide a foundation for the principled design of robust and versatile MOT systems suitable for real-world deployment.