SOTAVerified

Multiple Object Tracking

Multiple Object Tracking is the problem of automatically identifying multiple objects in a video and representing them as a set of trajectories with high accuracy.

Source: SOT for MOT

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–10 of 318 papers

TitleStatusHype
Learning better representations for crowded pedestrians in offboard LiDAR-camera 3D tracking-by-detectionCode0
LiDAR MOT-DETR: A LiDAR-based Two-Stage Transformer for 3D Multiple Object Tracking—0
Using Cross-Domain Detection Loss to Infer Multi-Scale Information for Improved Tiny Head Tracking—0
Towards Accurate State Estimation: Kalman Filter Incorporating Motion Dynamics for 3D Multi-Object Tracking—0
History-Aware Transformation of ReID Features for Multiple Object TrackingCode1
OVTR: End-to-End Open-Vocabulary Multiple Object Tracking with TransformerCode2
INTACT: Inducing Noise Tolerance through Adversarial Curriculum Training for LiDAR-based Safety-Critical Perception and Autonomy—0
Heterogeneous Graph Transformer for Multiple Tiny Object Tracking in RGB-T VideosCode1
A2VIS: Amodal-Aware Approach to Video Instance Segmentation—0
Enhancing Thermal MOT: A Novel Box Association Method Leveraging Thermal Identity and Motion SimilarityCode0
Show:102550
← PrevPage 1 of 32Next →
All datasetsKITTI Test (Online Methods)BDD100K valBDD100K testCroHDSportsMOTKITTI Test (Offline Methods)RADIATEWaymo Open DatasetGMOT-40UA-DETRACYouTube-Hands

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1MOTIP (Deformable DETR, with SportsMOT val)HOTA75.2—Unverified
2TrackSSMHOTA74.4—Unverified
3MOTIP (Deformable DETR)HOTA71.9—Unverified