Multi-Feature Re-Identification Enhanced Dual Motion Modeling for Multi Small-Object Tracking
Open Access
Abstract
Multi Small-Object Tracking (MSOT) is crucial for drone inspection and intelligent monitoring, yet traditional Multiple-object Tracking (MOT) methods perform poorly in such scenarios. The reasons include the following: small targets have low resolution and sparse features, leading to high missed detection rates; frequent occlusion and motion blur in dense scenes cause trajectory interruption and identity switches. To address these issues, an MSOT method combining dual motion modeling and dynamic Region of Interest (ROI) detection is proposed. The dual motion framework integrates Kalman filtering and optical flow through dynamic weighting to optimize target state estimation. The Kalman filter-guided dynamic ROI mechanism, combined with multi-feature fusion, enables trajectory recovery when targets are lost. Experiments on the VisDrone-MOT and UAVDT datasets show that this method outperforms mainstream algorithms in core metrics such as MOTA and HOTA, with better trajectory continuity and identity consistency while maintaining good real-time performance.Keywords
Funding Information
- National Key Research and Development Program of China (2023YFB4502804)
This publication has 29 references indexed in Scilit:
- Visual multi-object tracking with re-identification and occlusion handling using labeled random finite setsPattern Recognition, 2024
- Heterogeneous Graph Transformer for Multiple Tiny Object Tracking in RGB-T VideosIEEE Transactions on Multimedia, 2024
- DroneMOT: Drone-based Multi-Object Tracking Considering Detection Difficulties and Simultaneous Moving of Drones and Objects2024
- BoostTrack: boosting the similarity measure and detection confidence for improved multiple object trackingMachine Vision and Applications, 2024
- Deep OC-Sort: Multi-Pedestrian Tracking by Adaptive Re-Identification2023
- Interaction-Aware Labeled Multi-Bernoulli FilterIEEE Transactions on Intelligent Transportation Systems, 2023
- Observation-Centric SORT: Rethinking SORT for Robust Multi-Object Tracking2023
- VisDrone-MOT2019: The Vision Meets Drone Multiple Object Tracking Challenge Results2019
- Simple online and realtime tracking with a deep association metric2017
- Simple online and realtime tracking2016