Motion State: A New Benchmark Multiple Object Tracking

arXiv:2312.17641v2 Announce Type: replace
Abstract: In the realm of video analysis, the field of multiple object tracking (MOT) assumes paramount importance, with the motion state of objects-whether static or dynamic relative to the ground-holding practical significance across diverse scenarios. However, the extant literature exhibits a notable dearth in the exploration of this aspect. Deep learning methodologies encounter challenges in accurately discerning object motion states, while conventional approaches reliant on comprehensive mathematical modeling may yield suboptimal tracking accuracy. To address these …

abstract analysis arxiv benchmark cs.cv diverse dynamic exploration however importance literature multiple object objects practical realm significance state tracking type video video analysis