Multi Object Tracking On Sportsmot

Métriques

AssA

DetA

HOTA

IDF1

MOTA

Résultats

Résultats de performance de divers modèles sur ce benchmark

Nom du modèle	AssA	DetA	HOTA	IDF1	MOTA	Paper Title	Repository
QDTrack	47.2	77.5	60.4	62.3	90.1	Quasi-Dense Similarity Learning for Multiple Object Tracking
ETTrack	62.1	88.8	74.3	74.5	96.8	ETTrack: Enhanced Temporal Motion Predictor for Multi-Object Tracking	-
ByteTrack	52.3	78.5	64.1	71.4	95.9	ByteTrack: Multi-Object Tracking by Associating Every Detection Box
Deep HM-SORT	72.7	88.3	80.1	85.2	96.6	Deep HM-SORT: Enhancing Multi-Object Tracking in Sports with Deep Features, Harmonic Mean, and Expansion IOU	-
CenterTrack	48.0	82.1	62.7	60.0	90.8	Tracking Objects as Points
MixSort-OC	62.0	88.5	74.1	74.4	96.5	SportsMOT: A Large Multi-Object Tracking Dataset in Multiple Sports Scenes
MixSort-Byte	54.8	78.8	65.7	74.1	96.2	SportsMOT: A Large Multi-Object Tracking Dataset in Multiple Sports Scenes
OC-SORT	61.5	88.5	73.7	74.0	96.5	Observation-Centric SORT: Rethinking SORT for Robust Multi-Object Tracking
MeMOTR (Deformable-DETR)	57.8	82.0	68.8	69.9	90.2	MeMOTR: Long-Term Memory-Augmented Transformer for Multi-Object Tracking
TransTrack	57.5	82.7	68.9	71.5	92.6	TransTrack: Multiple Object Tracking with Transformer
MeMOTR	59.1	83.1	70.0	71.4	91.5	MeMOTR: Long-Term Memory-Augmented Transformer for Multi-Object Tracking
FairMOT	34.7	70.2	49.3	53.5	86.4	FairMOT: On the Fairness of Detection and Re-Identification in Multiple Object Tracking
GTR	45.9	64.8	54.5	55.8	67.9	Global Tracking Transformers
MambaMOT	58.6	86.7	71.3	71.1	94.9	MambaMOT: State-Space Model as Motion Predictor for Multi-Object Tracking	-
DeepMoveSORT	70.3	88.1	78.7	81.7	96.5	Engineering an Efficient Object Tracker for Non-Linear Motion	-
AED	70.1	89.4	79.1	81.8	97.1	Associate Everything Detected: Facilitating Tracking-by-Detection to the Unknown
MoveSORT	63.7	87.5	74.6	76.9	96.7	Beyond Kalman Filters: Deep Learning-Based Filters for Improved Object Tracking
Deep-EIoU	67.7	88.2	77.2	79.8	96.3	Iterative Scale-Up ExpansionIoU and Deep Features Association for Multi-Object Tracking in Sports	-
DeepEIoU + GTA	74.5	88.2	81.0	86.5	96.3	GTA: Global Tracklet Association for Multi-Object Tracking in Sports
MotionTrack	61.7	88.8	74.0	74.0	96.6	MotionTrack: Learning Motion Predictor for Multiple Object Tracking	-

0 of 20 row(s) selected.