mlsanityData to 7 Oct 2026

SFSORT: Scene Features-based Simple Online Real-Time Tracker

arXiv 2404.07553 ↗First posted 11 Apr 2024cs.CVcited by 2

Not on the map. Why?

Abstract (arXiv)

This paper introduces SFSORT, the world's fastest multi-object tracking system based on experiments conducted on MOT Challenge datasets. To achieve an accurate and computationally efficient tracker, this paper employs a tracking-by-detection method, following the online real-time tracking approach established in prior literature. By introducing a novel cost function called the Bounding Box Similarity Index, this work eliminates the Kalman Filter, leading to reduced computational requirements. Additionally, this paper demonstrates the impact of scene features on enhancing object-track association and improving track post-processing. Using a 2.2 GHz Intel Xeon CPU, the proposed method achieves an HOTA of 61.7% with a processing speed of 2242 Hz on the MOT17 dataset and an HOTA of 60.9% with a processing speed of 304 Hz on the MOT20 dataset. The tracker's source code, fine-tuned object detection model, and tutorials are available at urlhttps://github.com/gitmehrdad/SFSORT.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 43 of its 61 references that are in the map. First posted on arXiv 11 Apr 2024.

Builds on

43 of 61 references.

References · 43

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  2. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  3. 2016Wide Residual Networkscited by 3,057
  4. 2018YOLOv3: An Incremental Improvementcited by 2,652
  5. 2019FCOS: Fully Convolutional One-Stage Object Detectioncited by 1,573
  6. 2019Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regressioncited by 1,361
  7. 2017Cascade R-CNN: Delving into High Quality Object Detectioncited by 1,275
  8. 2021YOLOX: Exceeding YOLO Series in 2021cited by 1,064
  9. 2022YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectorscited by 978
  10. 2017Simple Online and Realtime Tracking with a Deep Association Metriccited by 908
  11. 2016Simple Online and Realtime Trackingcited by 862
  12. 2021ByteTrack: Multi-Object Tracking by Associating Every Detection Boxcited by 812
31 more
  1. 2019CenterNet: Keypoint Triplets for Object Detectioncited by 654
  2. 2020HOTA: A Higher Order Metric for Evaluating Multi-Object Trackingcited by 546
  3. 2019Distance-IoU Loss: Faster and Better Learning for Bounding Box Regressioncited by 454
  4. 2019Bag of Tricks and A Strong Baseline for Deep Person Re-identificationcited by 429
  5. 2020TransTrack: Multiple Object Tracking with Transformercited by 370
  6. 2020MOT20: A benchmark for multi object tracking in crowded scenescited by 355
  7. 2022Observation-Centric SORT: Rethinking SORT for Robust Multi-Object Trackingcited by 345
  8. 2022StrongSORT: Make DeepSORT Great Againcited by 213
  9. 2019Learning a Neural Solver for Multiple Object Trackingcited by 198
  10. 2018Acquisition of Localization Confidence for Accurate Object Detectioncited by 174
  11. 2020FastReID: A Pytorch Toolbox for General Instance Re-identificationcited by 164
  12. 2020MOTChallenge: A Benchmark for Single-Camera Multiple Target Trackingcited by 134
  13. 2023Deep OC-SORT: Multi-Pedestrian Tracking by Adaptive Re-Identificationcited by 130
  14. 2018Real-time Multiple People Tracking with Deeply Learned Candidate Selection and Person Re-Identificationcited by 96
  15. 2016POI: Multiple Object Tracking with High Performance Detection and Appearance Featurecited by 74
  16. 2017Deeply-Learned Part-Aligned Representations for Person Re-Identificationcited by 68
  17. 2022Hard to Track Objects with Irregular Motions and Similar Appearances? Make It Easier by Buffering the Matching Spacecited by 66
  18. 2023Hybrid-SORT: Weak Cues Matter for Online Multi-Object Trackingcited by 66
  19. 2022Unifying Short and Long-Term Tracking with Graph Hierarchiescited by 62
  20. 2022QDTrack: Quasi-Dense Similarity Learning for Appearance-Only Multiple Object Trackingcited by 61
  21. 2021Focal and Efficient IOU Loss for Accurate Bounding Box Regressioncited by 57
  22. 2021Learnable Graph Matching: Incorporating Graph Partitioning with Deep Feature Learning for Multiple Object Trackingcited by 47
  23. 2022GIAOTracker: A comprehensive framework for MCMOT with global information and optimizing strategies in VisDrone 2021cited by 46
  24. 2020Detecting Invisible Peoplecited by 31
  25. 2022SMILEtrack: SiMIlarity LEarning for Occlusion-Aware Multiple Object Trackingcited by 29
  26. 2021Semi-TCL: Semi-Supervised Track Contrastive Representation Learningcited by 23
  27. 2022Online Multi-Object Tracking with Unsupervised Re-Identification Learning and Occlusion Estimationcited by 16
  28. 2022Detection Recovery in Online Multi-Object Tracking with Sparse Graph Trackercited by 15
  29. 2023Localization-Guided Track: A Deep Association Multi-Object Tracking Framework Based on Localization Confidence of Detectionscited by 5
  30. 2023Multi-Object Tracking as Attention Mechanismcited by 3
  31. 2023BASE: Probably a Better Approach to Multi-Object Trackingcited by 1

Built on by

Cited by year

Citing papers per year: 2024: 2.

2 papers.

  1. 2024BoostTrack++: using tracklet information to detect more objects in multiple object trackingcited by 15
  2. 2024A Low-Computational Video Synopsis Framework with a Standard Datasetcited by 0