AI RESEARCH

Motion-Aware Transformer for Multi-Object Tracking

arXiv CS.CV

ArXi:2509.21715v3 Announce Type: replace Multi-object tracking (MOT) in videos remains challenging due to complex object motions and crowded scenes. Recent DETR-based frameworks offer end-to-end solutions but typically process detection and tracking queries jointly within a single Transformer Decoder layer, leading to conflicts and degraded association accuracy. We