mlsanityData to 14 Oct 2026

EarlyBird: Early-Fusion for Multi-View Tracking in the Bird's Eye View

arXiv 2310.13350 ↗First posted 20 Oct 2023cs.CVcited by 33

Not on the map. Why?

Abstract (arXiv)

Multi-view aggregation promises to overcome the occlusion and missed detection challenge in multi-object detection and tracking. Recent approaches in multi-view detection and 3D object detection made a huge performance leap by projecting all views to the ground plane and performing the detection in the Bird's Eye View (BEV). In this paper, we investigate if tracking in the BEV can also bring the next performance breakthrough in Multi-Target Multi-Camera (MTMC) tracking. Most current approaches in multi-view tracking perform the detection and tracking task in each view and use graph-based approaches to perform the association of the pedestrian across each view. This spatial association is already solved by detecting each pedestrian once in the BEV, leaving only the problem of temporal association. For the temporal association, we show how to learn strong Re-Identification (re-ID) features for each detection. The results show that early-fusion in the BEV achieves high accuracy for both detection and tracking. EarlyBird outperforms the state-of-the-art methods and improves the current state-of-the-art on Wildtrack by +4.6 MOTA and +5.6 IDF1.

Key facts. Cited by 33 papers in the map (data through 14 Oct 2026). Builds on 28 of its 55 references that are in the map. First posted on arXiv 20 Oct 2023.

Builds on

28 of 55 references.

References · 28

  1. 2017Mask R-CNNcited by 7,277
  2. 2020Supervised Contrastive Learningcited by 2,829
  3. 2018YOLOv3: An Incremental Improvementcited by 2,659
  4. 2022BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformerscited by 1,218
  5. 2019Objects as Pointscited by 1,175
  6. 2020Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3Dcited by 1,044
  7. 2017Simple Online and Realtime Tracking with a Deep Association Metriccited by 910
  8. 2016Performance Measures and a Data Set for Multi-Target, Multi-Camera Trackingcited by 692
  9. 2020FairMOT: On the Fairness of Detection and Re-Identification in Multiple Object Trackingcited by 490
  10. 2020Tracking Objects as Pointscited by 483
  11. 2019Tracking without bells and whistlescited by 382
  12. 2019Towards Real-Time Multi-Object Trackingcited by 311
16 more
  1. 2019MOTS: Multi-Object Tracking and Segmentationcited by 223
  2. 2017Detect to Track and Track to Detectcited by 131
  3. 2022Simple-BEV: What Really Matters for Multi-Sensor BEV Perception?cited by 128
  4. 2019CityFlow: A City-Scale Benchmark for Multi-Target Multi-Camera Vehicle Tracking and Re-Identificationcited by 119
  5. 2018Real-time Multiple People Tracking with Deeply Learned Candidate Selection and Person Re-Identificationcited by 96
  6. 2020Multiview Detection with Feature Perspective Transformationcited by 71
  7. 2021Multiview Detection with Shadow Transformer (and View-Coherent Data Augmentation)cited by 45
  8. 2022Simple Cues Lead to a Strong Multi-Object Trackercited by 44
  9. 2017Deep Occlusion Reasoning for Multi-Camera Multi-Target Detectioncited by 42
  10. 2017Deep Multi-camera People Detectioncited by 37
  11. 2021LMGP: Lifted Multicut Meets Geometry Projections for Multi-Camera Multi-Object Trackingcited by 34
  12. 2023ReST: A Reconfigurable Spatial-Temporal Graph Model for Multi-Camera Multi-Object Trackingcited by 24
  13. 2020Real-time 3D Deep Multi-Camera Trackingcited by 23
  14. 20223D Random Occlusion and Multi-Layer Projection for Deep Multi-Camera Pedestrian Localizationcited by 22
  15. 2021Lightweight Multi-Branch Network for Person Re-Identificationcited by 18
  16. 2022Synthehicle: Multi-Vehicle Multi-Camera Tracking in Virtual Citiescited by 12

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 8; 2025: 11; 2026: 14.

33 papers.

Sort:

By area:

  1. 2024Unleashing HyDRa: Hybrid Fusion, Depth Consistency and Radar for Unified 3D Perceptioncited by 35
  2. 2024Lifting Multi-View Detection and Tracking to the Bird's Eye Viewcited by 25
  3. 2024MCBLT: Multi-Camera Multi-Object 3D Tracking in Long Videoscited by 9
  4. 2024SpaRC: Sparse Radar-Camera Fusion for 3D Object Detectioncited by 8
  5. 2025Enhanced Multi-View Pedestrian Detection Using Probabilistic Occupancy Volumecited by 6
  6. 2025MITracker: Multi-View Integration for Visual Object Trackingcited by 5
  7. 2025DCHM: Depth-Consistent Human Modeling for Multiview Detectioncited by 5
  8. 2025MVTrajecter: Multi-View Pedestrian Tracking with Trajectory Motion Cost and Trajectory Appearance Costcited by 5
  9. 2024Resource-Efficient Multiview Perception: Integrating Semantic Masking with Masked Autoencoderscited by 2
  10. 20243D Multi-Object Tracking Employing MS-GLMB Filter for Autonomous Drivingcited by 2
  11. 2024MVUDA: Unsupervised Domain Adaptation for Multi-view Pedestrian Detectioncited by 2
  12. 2025Attention-Aware Multi-View Pedestrian Trackingcited by 2
21 more
  1. 2025One Graph to Track Them All: Dynamic GNNs for Single- and Multi-View Trackingcited by 2
  2. 2025Online 3D Multi-Camera Perception through Robust 2D Tracking and Depth-based Late Aggregationcited by 2
  3. 2026A Unified 3D Object Perception Framework for Real-Time Outside-In Multi-Camera Systemscited by 2
  4. 2024Spatial-Temporal Multi-Cuts for Online Multiple-Camera Vehicle Trackingcited by 1
  5. 2025SpaRC-AD: A Baseline for Radar-Camera Fusion in End-to-End Autonomous Drivingcited by 1
  6. 2025ATLASFusion: Aggregation Tracking with Location-Aware Sparse Fusion for Robust Spatio-Temporal Multi-View Pedestrian Trackingcited by 1
  7. 2025Multi Camera Connected Vision System with Multi View Analytics: A Comprehensive Surveycited by 1
  8. 20253D Ground Truth Reconstruction from Multi-Camera Annotations Using UKFcited by 0
  9. 2026Model Optimization for Multi-Camera 3D Detection and Trackingcited by 0
  10. 2026ModTrack: Sensor-Agnostic Multi-View Tracking via Identity-Informed PHD Filtering with Covariance Propagationcited by 0
  11. 2026LRC-WeatherNet: LiDAR, RADAR, and Camera Fusion Network for Real-time Weather-type Classification in Autonomous Drivingcited by 0
  12. 2026SynMVCrowd: A Large Synthetic Benchmark for Multi-view Crowd Counting and Localizationcited by 0
  13. 2026Efficient Online 3D Multi-Camera Multi-Object Tracking and Pose Estimationcited by 0
  14. 2026Multi-view Crowd Tracking Transformer with View-Ground Interactions Under Large Real-World Scenescited by 0
  15. 2026ARGUSTRACK: A Multi-View Annotation System for Multi-Object Trackingcited by 0
  16. 2026MVDGC: Joint 3D and 2D Multi-view Pedestrian Detection via Dual Geometric Constraintscited by 0
  17. 2026Syn2RealTrack: Bridging the Gap Between Synthetic and Real-World Datasets for Online Multi-View Multi-Target Trackingcited by 0
  18. 2026GRACE: Geometry- and Ray-Aware Camera-Efficient Multi-View Pedestrian Trackingcited by 0
  19. 2026Online Multi-Camera 3D Tracking via ID Prediction over Recurrent Sparse Queriescited by 0
  20. 2026Beyond the Survey: A Systematic Empirical Study of Detection and Association in Visual MOTcited by 0
  21. 2026CAT-Free: Multi-View Pedestrian Localization without Calibration, Annotations, or Target-Scene Training via Adaptive Geometric Filteringcited by 0