Intra- and inter-instance Location Correlation Network for human–object interaction detection

Engineering Applications of Artificial Intelligence, Год журнала: 2025, Номер 142, С. 109942 - 109942

Опубликована: Янв. 5, 2025

Язык: Английский

View adaptive multi-object tracking method based on depth relationship cues

Complex & Intelligent Systems, Год журнала: 2025, Номер 11(2)

Опубликована: Янв. 15, 2025

Multi-object tracking (MOT) tasks face challenges from multiple perception views due to the diversity of application scenarios. Different (front-view and top-view) have different imaging data distribution characteristics, but current MOT methods do not consider these differences only adopt a unified association strategy deal with various occlusion situations. This paper proposed View Adaptive Multi-Object Tracking Method Based on Depth Relationship Cues (ViewTrack) enable adapt scene's dynamic changes. Firstly, based exploiting depth relationships between objects by using position information bounding box, view-type recognition method relationship cues (VTRM) is perceive changes view within scene. Secondly, adjusting interval partitioning in adaptive for tracklet sets detection (VAPM) achieve sparse decomposition occluded scenes. Then, combining pedestrian displacement Intersection over Union (IoU), modulated (DMIoU) improve accuracy boxes. Finally, comparison results 12 representative demonstrate that ViewTrack outperforms metrics benchmark datasets. The code available at https://github.com/Hamor404/ViewTrack .

Язык: Английский

Advanced Early Warning System for Ship–Bridge Collisions Using Multisource Data Fusion and Maneuverability Assessment

ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems Part A Civil Engineering, Год журнала: 2025, Номер 11(3)

Опубликована: Апрель 23, 2025

Язык: Английский