MVDGC unifies BEV and image-view pedestrian localization into one task via 3D cylindrical queries that enforce dual geometric constraints between views.
hub
BoT-SORT: Robust Associations Multi-Pedestrian Tracking
29 Pith papers cite this work, alongside 310 external citations. Polarity classification is still indexing.
29
Pith papers citing it
310
external citations · Pith
abstract
The goal of multi-object tracking (MOT) is detecting and tracking all the objects in a scene, while keeping a unique identifier for each object. In this paper, we present a new robust state-of-the-art tracker, which can combine the advantages of motion and appearance information, along with camera-motion compensation, and a more accurate Kalman filter state vector. Our new trackers BoT-SORT, and BoT-SORT-ReID rank first in the datasets of MOTChallenge [29, 11] on both MOT17 and MOT20 test sets, in terms of all the main MOT metrics: MOTA, IDF1, and HOTA. For MOT17: 80.5 MOTA, 80.2 IDF1, and 65.0 HOTA are achieved. The source code and the pre-trained models are available at https://github.com/NirAharon/BOT-SORT
hub tools
citation-role summary
background 3
method 1
citation-polarity summary
representative citing papers
CityOS is an edge runtime that enforces a three-tier privacy API for urban sensors: local raw data, differentially private single-location stats, and cross-location aggregates with per-user budgets enforced on devices.
CROWD is a new global dataset of 51,753 continuous urban dashcam segments spanning over 20,000 hours from 238 countries, with manual labels and automated object detections for routine driving analysis.
A closed-set wasp-tracking benchmark shows current MOT trackers lose identities over thousands of frames even with perfect detections.
CylindTrack improves identity preservation in panoramic multi-object tracking by combining depth-temporal trajectory modeling, spherical spatio-temporal consistency learning, and topology-aware cylindrical motion prediction.
Presents a new egocentric HRI dataset and reports that an optimized tracking pipeline with ReID reduces identity switches by 49% over baseline.
GateMOT proposes Q-Gated Attention to enable linear-complexity, spatially aware attention for state-of-the-art dense object tracking on benchmarks like BEE24.
ART-Track is a motion-driven multi-object tracker that reduces identity switches in low-quality microgravity videos of model organisms by combining multi-model motion estimation, state-driven association, and uncertainty-adaptive cue fusion.
WildLIFT lifts monocular drone video to 3D for species-agnostic wildlife detection, tracking, and viewpoint analysis by integrating scene geometry with open-vocabulary segmentation.
SPL unifies unsupervised and sparsely-supervised 3D object detection via semantic pseudo-labeling that produces bounding boxes and point labels, followed by memory-based prototype learning that mines features from both labeled and unlabeled data.
EASE-MCVT is a distributed edge-assisted multi-camera vehicle tracking framework that achieves real-time performance and competitive accuracy on public datasets through edge processing and server-side optimizations.
OmniTrack++ improves omnidirectional multi-object tracking with trajectory feedback through DynamicSSM stabilization, FlexiTrack instances, ExpertTrack Memory with Mixture-of-Experts, and adaptive Tracklet Management, achieving SOTA HOTA gains on JRDB and new EmboTrack benchmark.
Introduces TimberVision dataset and multi-task framework for log-component segmentation, detection, and tracking in forestry operations using RGB images.
PS-Track sets a new state-of-the-art for point-supervised multi-object tracking by converting point seeds into temporally consistent pseudo-labels via Temporal-Feedback Prompting, Point-Excited Wavelet Attention, and Uncertainty-Guided Gaussian Learning.
Edge-TSR shows benchmark evaluations overestimate real-world edge inference performance by 20-30% and uses temporal stabilization to recover up to 10.16% classification accuracy in sustained roadside perception deployments.
IMPose introduces dual-level (keypoint and instance) correction propagation with a trajectory bank to turn sparse annotations into dense multi-person pose trajectories in videos.
SMAC introduces a spatial-modal fusion backbone and adaptive collapse network for multimodal MOT, reporting 63.31 HOTA and 79.21 MOTA on UniRTL RNT modality.
A three-stage pipeline applies YOLO11 detection, SAM segmentation, and persona-scaffolded adversarial chain-of-thought prompting on Qwen3-VL to monitor construction safety violations, reporting a 12% precision gain from the prompting method in an informal review.
SAMOFT improves multi-object tracking by using SAM segmentation and optical flow for pixel-level motion matching, flexible centroid correction, and training-free motion pattern fixes on top of standard Kalman and ReID baselines.
TCMP achieves SOTA MOT metrics (HOTA 63.4%, IDF1 65.0%, AssA 49.1%) with 0.014x parameters and 0.05x FLOPs of the previous best method by using a simple dilated TCN regressor.
HyperSSM integrates hypergraphs and state space models to let correlated objects mutually refine motion estimates, stabilizing trajectories under noise and occlusion for state-of-the-art multi-object tracking.
The paper derives an occlusion-aware multi-object tracking method that assigns each object an expected detection probability over the reduced Palm density within a multi-Bernoulli mixture filter.
NOOUGAT unifies online and offline multi-object tracking with a GNN that processes non-overlapping subclips fused by an Autoregressive Long-term Tracking layer, reporting SOTA gains on DanceTrack, SportsMOT, and MOT20.
Zero-shot VLM semantic descriptions achieve re-identification retrieval performance comparable to a supervised CNN baseline in autonomous driving but encounter attribute inconsistency across viewpoints.
citing papers explorer
-
MVDGC: Joint 3D and 2D Multi-view Pedestrian Detection via Dual Geometric Constraints
MVDGC unifies BEV and image-view pedestrian localization into one task via 3D cylindrical queries that enforce dual geometric constraints between views.
-
CityOS: Privacy Architecture for Urban Sensing
CityOS is an edge runtime that enforces a three-tier privacy API for urban sensors: local raw data, differentially private single-location stats, and cross-location aggregates with per-user budgets enforced on devices.
-
A global dataset of continuous urban dashcam driving
CROWD is a new global dataset of 51,753 continuous urban dashcam segments spanning over 20,000 hours from 238 countries, with manual labels and automated object detections for routine driving analysis.
-
WaspMOT: A Benchmark for Long-Term Multi-Object Tracking of Trichogramma Wasps
A closed-set wasp-tracking benchmark shows current MOT trackers lose identities over thousands of frames even with perfect detections.
-
CylindTrack: Depth-Aware Cylindrical Motion Modeling for Panoramic Multi-Object Tracking
CylindTrack improves identity preservation in panoramic multi-object tracking by combining depth-temporal trajectory modeling, spherical spatio-temporal consistency learning, and topology-aware cylindrical motion prediction.
-
Face versus Body Tracking for Human-Robot Interaction: An Egocentric Dataset
Presents a new egocentric HRI dataset and reports that an optimized tracking pipeline with ReID reduces identity switches by 49% over baseline.
-
GateMOT: Q-Gated Attention for Dense Object Tracking
GateMOT proposes Q-Gated Attention to enable linear-complexity, spatially aware attention for state-of-the-art dense object tracking on benchmarks like BEE24.
-
Motion-Driven Multi-Object Tracking of Model Organisms in Space Science Experiments
ART-Track is a motion-driven multi-object tracker that reduces identity switches in low-quality microgravity videos of model organisms by combining multi-model motion estimation, state-driven association, and uncertainty-adaptive cue fusion.
-
WildLIFT: Lifting monocular drone video to 3D for species-agnostic wildlife monitoring
WildLIFT lifts monocular drone video to 3D for species-agnostic wildlife detection, tracking, and viewpoint analysis by integrating scene geometry with open-vocabulary segmentation.
-
Unified Unsupervised and Sparsely-Supervised 3D Object Detection by Semantic Pseudo-Labeling and Prototype Learning
SPL unifies unsupervised and sparsely-supervised 3D object detection via semantic pseudo-labeling that produces bounding boxes and point labels, followed by memory-based prototype learning that mines features from both labeled and unlabeled data.
-
Edge Assisted Multi-Camera Vehicle Tracking Framework for Real-Time and Scalable Deployment
EASE-MCVT is a distributed edge-assisted multi-camera vehicle tracking framework that achieves real-time performance and competitive accuracy on public datasets through edge processing and server-side optimizations.
-
OmniTrack++: Omnidirectional Multi-Object Tracking by Learning Large-FoV Trajectory Feedback
OmniTrack++ improves omnidirectional multi-object tracking with trajectory feedback through DynamicSSM stabilization, FlexiTrack instances, ExpertTrack Memory with Mixture-of-Experts, and adaptive Tracklet Management, achieving SOTA HOTA gains on JRDB and new EmboTrack benchmark.
-
TimberVision: A Multi-Task Dataset and Framework for Log-Component Segmentation and Tracking in Autonomous Forestry Operations
Introduces TimberVision dataset and multi-task framework for log-component segmentation, detection, and tracking in forestry operations using RGB images.
-
PS-MOT: Cultivating Instance Awareness from Point Seeds for Multi-Object Tracking
PS-Track sets a new state-of-the-art for point-supervised multi-object tracking by converting point seeds into temporally consistent pseudo-labels via Temporal-Feedback Prompting, Point-Excited Wavelet Attention, and Uncertainty-Guided Gaussian Learning.
-
Beyond Benchmarks: Continuous Edge Inference for Fine-Grained Roadside Perception
Edge-TSR shows benchmark evaluations overestimate real-world edge inference performance by 20-30% and uses temporal stabilization to recover up to 10.16% classification accuracy in sustained roadside perception deployments.
-
IMPose: Interactive Multi-person Pose Estimation with Dynamic Correction Propagation
IMPose introduces dual-level (keypoint and instance) correction propagation with a trajectory bank to turn sparse annotations into dense multi-person pose trajectories in videos.
-
SMAC: Spatial-Modal Joint Modeling and Adaptive Representation Collapse for Multimodal Object Tracking
SMAC introduces a spatial-modal fusion backbone and adaptive collapse network for multimodal MOT, reporting 63.31 HOTA and 79.21 MOTA on UniRTL RNT modality.
-
Passive Construction Site Safety Monitoring via Persona-Scaffolded Adversarial Chain-of-Thought VLM Verification
A three-stage pipeline applies YOLO11 detection, SAM segmentation, and persona-scaffolded adversarial chain-of-thought prompting on Qwen3-VL to monitor construction safety violations, reporting a 12% precision gain from the prompting method in an informal review.
-
SAMOFT: Robust Multi-Object Tracking via Region and Flow
SAMOFT improves multi-object tracking by using SAM segmentation and optical flow for pixel-level motion matching, flexible centroid correction, and training-free motion pattern fixes on top of standard Kalman and ReID baselines.
-
Time-series Meets Complex Motion Modeling: Robust and Computational-effective Motion Predictor for Multi-object Tracking
TCMP achieves SOTA MOT metrics (HOTA 63.4%, IDF1 65.0%, AssA 49.1%) with 0.014x parameters and 0.05x FLOPs of the previous best method by using a simple dilated TCN regressor.
-
Hypergraph-State Collaborative Reasoning for Multi-Object Tracking
HyperSSM integrates hypergraphs and state space models to let correlated objects mutually refine motion estimates, stabilizing trajectories under noise and occlusion for state-of-the-art multi-object tracking.
-
Occlusion-Aware Multi-Object Tracking via Expected Probability of Detection
The paper derives an occlusion-aware multi-object tracking method that assigns each object an expected detection probability over the reduced Palm density within a multi-Bernoulli mixture filter.
-
NOOUGAT: Towards Unified Online and Offline Multi-Object Tracking
NOOUGAT unifies online and offline multi-object tracking with a GNN that processes non-overlapping subclips fused by an Autoregressive Long-term Tracking layer, reporting SOTA gains on DanceTrack, SportsMOT, and MOT20.
-
Zero-Shot Semantic Re-Identification for Autonomous Driving: A VLM Baseline Study
Zero-shot VLM semantic descriptions achieve re-identification retrieval performance comparable to a supervised CNN baseline in autonomous driving but encounter attribute inconsistency across viewpoints.
-
Multi-Object Tracking Consistently Improves Wildlife Inference
Applying multi-object tracking to fuse softmax probabilities across frames in camera trap data yields weighted F1-score gains of 5.1%, 3.1%, and 2.0% over standalone classifiers on three datasets.
-
OpenPodcar2: a robust, ROS2 vehicle for self-driving research
OpenPodcar2 is a low-cost, open-source ROS2 autonomous vehicle platform built from a mobility scooter, with hardware build instructions, Gazebo simulation, and nav2-based planning for research and limited deployment.
-
EgoLive: A Large-Scale Egocentric Dataset from Real-World Human Tasks
EgoLive is presented as the largest open-source annotated egocentric dataset for real-world task-oriented human routines, captured with a custom head-mounted device and multi-modal annotations exclusively in unconstrained environments.
-
SAMIDARE: Advanced Tracking-by-Segmentation for Dense Scenarios
SAMIDARE improves segmentation-based multi-object tracking in dense sports videos, gaining 2.5 HOTA and 4.2 IDF1 over baseline on SportsMOT validation through adaptive mask control and state-aware association.
-
Trajectory Prediction for Autonomous Driving: Progress, Limitations, and Future Directions
A survey of trajectory prediction techniques for autonomous vehicles that proposes a taxonomy, overviews the prediction pipeline, and highlights remaining research gaps.