Multi Object Tracking On Mot20 1

평가 지표

IDF1

MOTA

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	IDF1	MOTA	Paper Title	Repository
TransCenter	57.9	72.4	TransCenter: Transformers with Dense Representations for Multiple-Object Tracking
BoT-SORT	77.5	77.8	BoT-SORT: Robust Associations Multi-Pedestrian Tracking
AdapTrack	80.7	75.0	AdapTrack: Adaptive Thresholding-Based Matching For Multi-object Tracking
BoostTrack+	81.5	77.2	BoostTrack: boosting the similarity measure and detection confidence for improved multiple object tracking
SimpleTrack	70.2	72.6	SimpleTrack: Rethinking and Improving the JDE Approach for Multi-Object Tracking
UCMCTrack	77.4	-	UCMCTrack: Multi-Object Tracking with Uniform Camera Motion Compensation
STGT	75.2	77.5	TransMOT: Spatial-Temporal Graph Transformer for Multiple Object Tracking	-
FCG	69.7	68.0	Multiple Object Tracking from appearance by hierarchically clustering tracklets
GSDT	-	67.1	Joint Object Detection and Multi-Object Tracking with Graph Neural Networks
IMM-JHSE	74.64	72.82	One Homography is All You Need: IMM-based Joint Homography and Multiple Object State Estimation
CMTrack	79.9	76.2	A Confidence-Aware Matching Strategy For Generalized Multi-Object Tracking
Deep OC-SORT	79.2	75.6	Deep OC-SORT: Multi-Pedestrian Tracking by Adaptive Re-Identification
MPNTrack	59.1	57.6	Learning a Neural Solver for Multiple Object Tracking
BoostTrack++	82	77.7	BoostTrack++: using tracklet information to detect more objects in multiple object tracking
ByteTrack	75.2	77.8	ByteTrack: Multi-Object Tracking by Associating Every Detection Box
OUTrack_fm	69.4	68.5	Online Multi-Object Tracking with Unsupervised Re-Identification Learning and Occlusion Estimation	-
SparseTrack	77.3	78.2	SparseTrack: Multi-Object Tracking by Performing Scene Decomposition based on Pseudo-Depth
SGT	70.6	72.8	Detection Recovery in Online Multi-Object Tracking with Sparse Graph Tracker
HopTrack[Embedded GPU]	-	45.6	HopTrack: A Real-time Multi-Object Tracking System for Embedded Devices
LMOT	61.1	59.1	LMOT: Efficient Light-Weight Detection and Tracking in Crowds

0 of 26 row(s) selected.