arXiv:2510.24410v1 Announce Type: new 
Abstract: This paper proposes a visual multi-object tracking method that jointly employs stochastic and deterministic mechanisms to ensure identifier consistency for unknown and time-varying target numbers under nonlinear dynamics. A stochastic particle filter addresses nonlinear dynamics and non-Gaussian noise, with support from particle swarm optimization (PSO) to guide particles toward state distribution modes and mitigate divergence through proposed fitness measures incorporating motion consistency, appearance similarity, and social-interaction cues with neighboring targets. Deterministic association further enforces identifier consistency via a proposed cost matrix incorporating spatial consistency between particles and current detections, detection confidences, and track penalties. Subsequently, a novel scheme is proposed for the smooth updating of target states while preserving their identities, particularly for weak tracks during interactions with other targets and prolonged occlusions. Moreover, velocity regression over past states provides trend-seed velocities, enhancing particle sampling and state updates. The proposed tracker is designed to operate flexibly for both pre-recorded videos and camera live streams, where future frames are unavailable. Experimental results confirm superior performance compared to state-of-the-art trackers. The source-code reference implementations of both the proposed method and compared-trackers are provided on GitHub: https://github.com/SDU-VelKoTek/GenTrack2

تقدم ورقة جديدة طريقة مبتكرة لتتبع الكائنات المتعددة بصريًا تجمع بين الأساليب العشوائية والحتمية للحفاظ على اتساق المعرفات لأعداد الأهداف المتغيرة. هذا مهم لأنه يعالج التحديات في التتبع تحت الديناميات غير الخطية والضوضاء، مما قد يحسن التطبيقات في الروبوتات والمراقبة.

Un nuevo artículo presenta un método innovador de seguimiento visual de múltiples objetos que combina enfoques estocásticos y deterministas para mantener la consistencia de los identificadores para números de objetivos variables. Esto es significativo porque aborda los desafíos del seguimiento bajo dinámicas no lineales y ruido, lo que podría mejorar aplicaciones en robótica y vigilancia.

Un nouvel article présente une méthode innovante de suivi visuel multi-objets qui combine des approches stochastiques et déterministes pour maintenir la cohérence des identifiants pour des nombres cibles variables. Cela est important car cela répond aux défis du suivi sous des dynamiques non linéaires et du bruit, ce qui pourrait améliorer les applications en robotique et en surveillance.

A new paper introduces an innovative visual multi-object tracking method that combines stochastic and deterministic approaches to maintain identifier consistency for varying target numbers. This is significant because it addresses challenges in tracking under nonlinear dynamics and noise, potentially improving applications in robotics and surveillance.

A Hybrid Approach for Visual Multi-Object Tracking

Was this article worth reading? Share it

LucidQuery AI

Open Source Surveillance

Video Toolkit

DeepTracker

Https

Metaflow AI

Ready to build your own newsroom?