Multi Object Tracking On Mot17

Metrics

MOTA

Results

Performance results of various models on this benchmark

		Paper Title	Repository
FeatureSORT	81.2	FeatureSORT: Essential Features for Effective Tracking	-
SMILEtrack	81.06	SMILEtrack: SiMIlarity LEarning for Occlusion-Aware Multiple Object Tracking
SparseTrack	81.0	SparseTrack: Multi-Object Tracking by Performing Scene Decomposition based on Pseudo-Depth
CMTrack	80.7	A Confidence-Aware Matching Strategy For Generalized Multi-Object Tracking	-
BoostTrack++	80.7	BoostTrack++: using tracklet information to detect more objects in multiple object tracking
BoostTrack+	80.6	BoostTrack: boosting the similarity measure and detection confidence for improved multiple object tracking	-
BoT-SORT	80.5	BoT-SORT: Robust Associations Multi-Pedestrian Tracking
UCMCTrack	80.5	UCMCTrack: Multi-Object Tracking with Uniform Camera Motion Compensation
ByteTrack	80.3	ByteTrack: Multi-Object Tracking by Associating Every Detection Box
AdapTrack	79.9	AdapTrack: Adaptive Thresholding-Based Matching For Multi-object Tracking	-
StrongSORT	79.6	StrongSORT: Make DeepSORT Great Again
IMM-JHSE	79.54	One Homography is All You Need: IMM-based Joint Homography and Multiple Object State Estimation
Deep OC-SORT	79.4	Deep OC-SORT: Multi-Pedestrian Tracking by Adaptive Re-Identification
SFSORT	78.8	SFSORT: Scene Features-based Simple Online Real-Time Tracker
C-TWiX	78.1	Learning Data Association for Multi-Object Tracking using Only Coordinates
OC-SORT	78.0	Observation-Centric SORT: Rethinking SORT for Robust Multi-Object Tracking
Unicorn	77.2	Towards Grand Unification of Object Tracking
STGT	76.7	TransMOT: Spatial-Temporal Graph Transformer for Multiple Object Tracking	-
FCG	76.7	Multiple Object Tracking from appearance by hierarchically clustering tracklets
SGT	76.4	Detection Recovery in Online Multi-Object Tracking with Sparse Graph Tracker

0 of 44 row(s) selected.

Command Palette

Multi Object Tracking On Mot17

Metrics

Results