SAMIDARE: Advanced Tracking-by-Segmentation for Dense Scenarios

arXiv cs.CV / 4/27/2026

📰 NewsTools & Practical UsageModels & Research

共有:

Key Points

The paper introduces SAMIDARE, an advanced segmentation-based multi-object tracking framework designed to handle dense, crowded sports scenarios where mask errors and ID switches are common.
SAMIDARE improves upon SAM2MOT by adding density-aware mask re-generation and selective memory updates to better preserve target feature integrity under challenging visual conditions.
It also uses state-aware association and new track initialization to increase robustness during mutual occlusions and frequent frame-out events.
On the SportsMOT dataset, SAMIDARE achieves state-of-the-art results, improving over the baseline by 2.5 HOTA and 4.2 IDF1 points on the validation set.
The authors provide accompanying code via the project’s GitHub repository, enabling reproducibility and further development.

Abstract

Automated sports analysis demands robust multi-object tracking (MOT), yet segmentation-based methods often struggle with mask errors and ID switches in dense scenes. We propose SAMIDARE, a framework that enhances SAM2MOT for crowded scenes through three key components: (1) density-aware mask re-generation and (2) selective memory updates, both for adaptive mask control to preserve target feature integrity, and (3) state-aware association and new track initialization, which improves robustness under mutual occlusions and frequent frame-out events. Evaluated on the SportsMOT dataset, SAMIDARE achieves state-of-the-art performance, outperforming the baseline by 2.5 HOTA and 4.2 IDF1 points on the validation set. These results demonstrate that adaptive feature management using mask control and state-aware association provide a robust and efficient solution for dense sports tracking. Code is available at https://github.com/ZabuZabuZabu/SAMIDARE