mlsanityData to 13 Oct 2026

Learning Data Association for Multi-Object Tracking using Only Coordinates

arXiv 2403.08018 ↗First posted 12 Mar 2024cs.CVcited by 8

Not on the map. Why?

Abstract (arXiv)

We propose a novel Transformer-based module to address the data association problem for multi-object tracking. From detections obtained by a pretrained detector, this module uses only coordinates from bounding boxes to estimate an affinity score between pairs of tracks extracted from two distinct temporal windows. This module, named TWiX, is trained on sets of tracks with the objective of discriminating pairs of tracks coming from the same object from those which are not. Our module does not use the intersection over union measure, nor does it requires any motion priors or any camera motion compensation technique. By inserting TWiX within an online cascade matching pipeline, our tracker C-TWiX achieves state-of-the-art performance on the DanceTrack and KITTIMOT datasets, and gets competitive results on the MOT17 dataset. The code will be made available upon publication.

Key facts. Cited by 8 papers in the map (data through 13 Oct 2026). Builds on 44 of its 58 references that are in the map. First posted on arXiv 12 Mar 2024.

Builds on

44 of 58 references.

References · 44

  1. 2017Attention Is All You Needcited by 60,106
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,304
  3. 2014Adam: A Method for Stochastic Optimizationcited by 37,093
  4. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,043
  5. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,009
  6. 2018Representation Learning with Contrastive Predictive Codingcited by 7,712
  7. 2020End-to-End Object Detection with Transformerscited by 7,396
  8. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,730
  9. 2015FaceNet: A Unified Embedding for Face Recognition and Clusteringcited by 2,838
  10. 2019Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regressioncited by 1,362
  11. 2021YOLOX: Exceeding YOLO Series in 2021cited by 1,069
  12. 2017Simple Online and Realtime Tracking with a Deep Association Metriccited by 908
32 more
  1. 2016Simple Online and Realtime Trackingcited by 864
  2. 2021ByteTrack: Multi-Object Tracking by Associating Every Detection Boxcited by 816
  3. 2017In Defense of the Triplet Loss for Person Re-Identificationcited by 785
  4. 2016Performance Measures and a Data Set for Multi-Target, Multi-Camera Trackingcited by 687
  5. 2019CenterNet: Keypoint Triplets for Object Detectioncited by 654
  6. 2020HOTA: A Higher Order Metric for Evaluating Multi-Object Trackingcited by 549
  7. 2016MOT16: A Benchmark for Multi-Object Trackingcited by 532
  8. 2021TrackFormer: Multi-Object Tracking with Transformerscited by 516
  9. 2020Tracking Objects as Pointscited by 483
  10. 2019Distance-IoU Loss: Faster and Better Learning for Bounding Box Regressioncited by 455
  11. 2019Tracking without bells and whistlescited by 382
  12. 2021MOTR: End-to-End Multiple-Object Tracking with Transformercited by 380
  13. 2022BoT-SORT: Robust Associations Multi-Pedestrian Trackingcited by 317
  14. 2020Quasi-Dense Similarity Learning for Multiple Object Trackingcited by 220
  15. 2019Learning a Neural Solver for Multiple Object Trackingcited by 198
  16. 2020Transformer Networks for Trajectory Forecastingcited by 180
  17. 2021Learning to Track with Object Permanencecited by 118
  18. 2019Analyzing and Improving Representations with the Soft Nearest Neighbor Losscited by 95
  19. 2018Deep Affinity Network for Multiple Object Trackingcited by 82
  20. 2016POI: Multiple Object Tracking with High Performance Detection and Appearance Featurecited by 74
  21. 2022Hard to Track Objects with Irregular Motions and Similar Appearances? Make It Easier by Buffering the Matching Spacecited by 66
  22. 2021Do Different Tracking Tasks Require Different Appearance Models?cited by 62
  23. 2022Unifying Short and Long-Term Tracking with Graph Hierarchiescited by 62
  24. 2020Social NCE: Contrastive Learning of Socially-aware Motion Representationscited by 58
  25. 2023SparseTrack: Multi-Object Tracking by Performing Scene Decomposition based on Pseudo-Depthcited by 48
  26. 2023UCMCTrack: Multi-Object Tracking with Uniform Camera Motion Compensationcited by 44
  27. 2020Probabilistic Tracklet Scoring and Inpainting for Multiple Object Trackingcited by 36
  28. 2021Track without Appearance: Learn Box and Tracklet Embedding with Local and Global Motion Patterns for Vehicle Trackingcited by 26
  29. 2021Semi-TCL: Semi-Supervised Track Contrastive Representation Learningcited by 23
  30. 2021Looking Beyond Two Frames: End-to-End Multi-Object Tracking Using Spatial and Temporal Transformerscited by 21
  31. 2020Towards Good Practices in Self-supervised Representation Learningcited by 19
  32. 2022Object Permanence Emerges in a Random Walk along Memorycited by 18

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2024: 3; 2025: 4; 2026: 1.

8 papers.

  1. 2024FeatureSORT: Essential Features for Effective Trackingcited by 8
  2. 2025Deep Learning-Based Multi-Object Tracking: A Comprehensive Survey from Foundations to State-of-the-Artcited by 4
  3. 2025FastTracker: Real-Time and Accurate Visual Trackingcited by 4
  4. 2024Learning a Neural Association Network for Self-supervised Multi-Object Trackingcited by 1
  5. 2025Stable at Any Speed: Speed-Driven Multi-Object Tracking with Learnable Kalman Filteringcited by 1
  6. 2024Tracking by Detection and Query: An Efficient End-to-End Framework for Multi-Object Trackingcited by 0
  7. 2025Learning Association via Track-Detection Matching for Multi-Object Trackingcited by 0
  8. 2026Towards Automatic Video Annotation with ASH: Zero-Shot Open-Vocabulary Multi-Object Tracking and Segmentationcited by 0