2D MOT 2015 Results
Click on a measure to sort the table accordingly. See below for a more detailed description.
| Tracker | MOTA | IDF1 | MOTP | MT | ML | FP | FN | Recall | Precision | FAF | ID Sw. | Frag | Hz | |
| TO 1.  |
25.7 ±13.5 |
32.7 ±11.3 | 72.2 | 31 (4.3) | 414 (57.4) | 4,779 | 40,511 | 34.1 | 81.4 | 0.8 | 383 (11.2) | 600 (17.6) | 5.0 | |
| S. Manen, R. Timofte, D. Dai, L. Gool. Leveraging single for multi-target tracking using a novel trajectory overlap affinity measure. In 2016 IEEE Winter Conference on Applications of Computer Vision (WACV), 2016. | ||||||||||||||
| UN_DAM 2.  |
29.7 ±12.3 |
41.4 ±10.2 | 71.4 | 66 (9.2) | 360 (49.9) | 7,610 | 35,269 | 42.6 | 77.5 | 1.3 | 318 (7.5) | 674 (15.8) | 20.7 | |
| Multi Object Tracking using Deep Structural Cost Minimization in Data Association | ||||||||||||||
| SegTrack 3. |
22.5 ±0.0 |
31.5 ±0.0 | 71.7 | 42 (5.8) | 461 (63.9) | 7,890 | 39,020 | 36.5 | 74.0 | 1.4 | 697 (19.1) | 737 (20.2) | 0.2 | |
| A. Milan, L. Leal-Taixé, K. Schindler, I. Reid. Joint Tracking and Segmentation of Multiple Targets. In CVPR, 2015. | ||||||||||||||
| LINF1 4. |
24.5 ±15.4 |
34.8 ±12.8 | 71.3 | 40 (5.5) | 466 (64.6) | 5,864 | 40,207 | 34.6 | 78.4 | 1.0 | 298 (8.6) | 744 (21.5) | 7.5 | |
| L. Fagot-Bouquet, R. Audigier, Y. Dhome, F. Lerasle. Improving Multi-Frame Data Association with Sparse Representations for Robust Near-Online Multi-Object Tracking. In ECCV, 2016. | ||||||||||||||
| HAM_SADF 5. |
25.2 ±13.4 |
37.8 ±12.2 | 71.4 | 41 (5.7) | 420 (58.3) | 7,330 | 38,275 | 37.7 | 76.0 | 1.3 | 357 (9.5) | 745 (19.8) | 18.7 | |
| Y. Yoon, A. Boragule, Y. Song, K. Yoon, M. Jeon. Online Multi-Object Tracking with Historical Appearance Matching and Scene Adaptive Detection Filtering. In IEEE AVSS, 2018. | ||||||||||||||
| TFMOT 6. |
23.8 ±12.0 |
32.3 ±11.7 | 71.3 | 35 (4.9) | 447 (62.0) | 4,533 | 41,873 | 31.8 | 81.2 | 0.8 | 404 (12.7) | 792 (24.9) | 11.3 | |
| M. Abhijeet Boragule. Joint Cost Minimization for Multi-Object Tracking. In 2017 IEEE International Conference on Advanced Vide and Signale Based Surveillance, 2017. | ||||||||||||||
| DCO_X 7. |
19.6 ±0.0 |
31.5 ±0.0 | 71.4 | 37 (5.1) | 396 (54.9) | 10,652 | 38,232 | 37.8 | 68.5 | 1.8 | 521 (13.8) | 819 (21.7) | 0.3 | |
| A. Milan, K. Schindler, S. Roth. Multi-Target Tracking by Discrete-Continuous Energy Minimization. In IEEE PAMI, 2016. | ||||||||||||||
| NOMT 8. |
33.7 ±16.2 |
44.6 ±14.3 | 71.9 | 88 (12.2) | 317 (44.0) | 7,762 | 32,547 | 47.0 | 78.8 | 1.3 | 442 (9.4) | 823 (17.5) | 11.5 | |
| W. Choi. Near-Online Multi-target Tracking with Aggregated Local Flow Descriptor. In ICCV, 2015. | ||||||||||||||
| TC_SIAMESE 9. |
20.2 ±12.6 |
32.6 ±13.4 | 71.1 | 19 (2.6) | 487 (67.5) | 6,127 | 42,596 | 30.7 | 75.5 | 1.1 | 294 (9.6) | 825 (26.9) | 13.0 | |
| Y. Yoon, Y. Song, K. Yoon, M. Jeon. Online Multiple-Object Tracking using Selective Deep Appearance Matching. In IEEE/IEIE The International Conference on Consumer Electronics (ICCE) Asia, 2018. | ||||||||||||||
| MHT_DAM 10. |
32.4 ±0.0 |
45.3 ±0.0 | 71.8 | 115 (16.0) | 316 (43.8) | 9,064 | 32,060 | 47.8 | 76.4 | 1.6 | 435 (9.1) | 826 (17.3) | 0.7 | |
| C. Kim, F. Li, A. Ciptadi, J. Rehg. Multiple Hypothesis Tracking Revisited. In ICCV, 2015. | ||||||||||||||
| Tracker | MOTA | IDF1 | MOTP | MT | ML | FP | FN | Recall | Precision | FAF | ID Sw. | Frag | Hz | |
| LP_SSVM 11. |
25.2 ±13.7 |
34.0 ±9.7 | 71.7 | 42 (5.8) | 382 (53.0) | 8,369 | 36,932 | 39.9 | 74.5 | 1.4 | 646 (16.2) | 849 (21.3) | 41.3 | |
| S. Wang, C. Fowlkes. Learning Optimal Parameters for Multi-target Tracking with Contextual Interactions. In International Journal of Computer Vision, 2016. | ||||||||||||||
| JPDA_m 12. |
23.8 ±0.0 |
33.8 ±0.0 | 68.2 | 36 (5.0) | 419 (58.1) | 6,373 | 40,084 | 34.8 | 77.0 | 1.1 | 365 (10.5) | 869 (25.0) | 32.6 | |
| H. Rezatofighi, A. Milan, Z. Zhang, Q. Shi, A. Dick, I. Reid. Joint Probabilistic Data Association Revisited. In ICCV, 2015. | ||||||||||||||
| MPNTrack 13. |
51.5 ±10.8 |
58.6 ±7.5 | 76.0 | 225 (31.2) | 187 (25.9) | 7,620 | 21,780 | 64.6 | 83.9 | 1.3 | 375 (5.8) | 872 (13.5) | 6.5 | |
| G. Brasó, L. Leal-Taixé. Learning a Neural Solver for Multiple Object Tracking. In CVPR, 2020. | ||||||||||||||
| SLTV15 14. |
27.6 ±14.4 |
40.3 ±11.5 | 71.4 | 52 (7.2) | 374 (51.9) | 6,581 | 37,566 | 38.9 | 78.4 | 1.1 | 358 (9.2) | 884 (22.7) | 20.9 | |
| Gwangju Institute of Science and Technology(GIST), Machine Learning and Vision Laboratory | ||||||||||||||
| CNNTCM 15. |
29.6 ±13.9 |
36.8 ±13.4 | 71.8 | 81 (11.2) | 317 (44.0) | 7,786 | 34,733 | 43.5 | 77.4 | 1.3 | 712 (16.4) | 943 (21.7) | 1.7 | |
| B. Wang, K. L. Chan, L. Wang, B. Shuai, Z. Zuo, T. Liu, G. Wang. Joint Learning of Convolutional Neural Networks and Temporally Constrained Metrics for Tracklet Association. In DeepVision Workshop (CVPR), 2016. | ||||||||||||||
| TSMLCDEnew 16. |
34.3 ±13.1 |
44.1 ±11.9 | 71.7 | 101 (14.0) | 284 (39.4) | 7,869 | 31,908 | 48.1 | 79.0 | 1.4 | 618 (12.9) | 959 (20.0) | 6.5 | |
| B. Wang, G. Wang, K. L. Chan, L. Wang. Tracklet Association by Online Target-Specific Metric Learning and Coherent Dynamics Estimation. In arXiv:1511.06654, 2015. | ||||||||||||||
| JointMC 17. |
35.6 ±17.5 |
45.1 ±9.8 | 71.9 | 167 (23.2) | 283 (39.3) | 10,580 | 28,508 | 53.6 | 75.7 | 1.8 | 457 (8.5) | 969 (18.1) | 0.6 | |
| M. Keuper, S. Tang, B. Andres, T. Brox, B. Schiele. Motion Segmentation amp; Multiple Object Tracking by Correlation Co-Clustering. In IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018. | ||||||||||||||
| MCF_PHD 18. |
29.9 ±20.0 |
38.2 ±13.4 | 71.7 | 86 (11.9) | 317 (44.0) | 8,892 | 33,529 | 45.4 | 75.8 | 1.5 | 656 (14.4) | 989 (21.8) | 12.2 | |
| N. Wojke, D. Paulus. Global data association for the Probability Hypothesis Density filter using network flows. In 2016 IEEE International Conference on Robotics and Automation, ICRA, 2016. | ||||||||||||||
| GSCR 19. |
15.8 ±10.8 |
27.9 ±10.1 | 69.4 | 13 (1.8) | 440 (61.0) | 7,597 | 43,633 | 29.0 | 70.1 | 1.3 | 514 (17.7) | 1,010 (34.8) | 28.1 | |
| L. Fagot-Bouquet, R. Audigier, Y. Dhome, F. Lerasle. Online multi-person tracking based on global sparse collaborative representations. In ICIP, 2015. | ||||||||||||||
| CEM 20. |
19.3 ±37.1 |
0.0 ±0.0 | 70.7 | 61 (8.5) | 335 (46.5) | 14,180 | 34,591 | 43.7 | 65.4 | 2.5 | 813 (18.6) | 1,023 (23.4) | 1.1 | |
| A. Milan, S. Roth, K. Schindler. Continuous Energy Minimization for Multitarget Tracking. In IEEE TPAMI, 2014. | ||||||||||||||
| Tracker | MOTA | IDF1 | MOTP | MT | ML | FP | FN | Recall | Precision | FAF | ID Sw. | Frag | Hz | |
| HAM_INTP15 21. |
28.6 ±13.8 |
41.4 ±10.0 | 71.1 | 72 (10.0) | 317 (44.0) | 7,485 | 35,910 | 41.6 | 77.3 | 1.3 | 460 (11.1) | 1,038 (25.0) | 18.7 | |
| Y. Yoon, A. Boragule, Y. Song, K. Yoon, M. Jeon. Online Multi-Object Tracking with Historical Appearance Matching and Scene Adaptive Detection Filtering. In IEEE AVSS, 2018. | ||||||||||||||
| Lif_T 22. |
52.5 ±0.0 |
60.0 ±0.0 | 76.3 | 244 (33.8) | 186 (25.8) | 6,837 | 21,610 | 64.8 | 85.3 | 1.2 | 730 (11.3) | 1,047 (16.2) | 1.5 | |
| Anonymous submission | ||||||||||||||
| MotiCon 23. |
23.1 ±16.4 |
29.4 ±9.8 | 70.9 | 34 (4.7) | 375 (52.0) | 10,404 | 35,844 | 41.7 | 71.1 | 1.8 | 1,018 (24.4) | 1,061 (25.5) | 1.4 | |
| L. Leal-Taixé, M. Fenzi, A. Kuznetsova, B. Rosenhahn, S. Savarese. Learning an image-based motion context for multiple people tracking. In CVPR, 2014. | ||||||||||||||
| LFNF 24. |
31.6 ±13.2 |
33.1 ±8.3 | 72.0 | 69 (9.6) | 301 (41.7) | 5,943 | 35,095 | 42.9 | 81.6 | 1.0 | 961 (22.4) | 1,106 (25.8) | 4.0 | |
| Sheng H, Hao L, Chen J, et al. Robust Local Effective Matching Model for Multi-Target Tracking. In PCM, 2017 | ||||||||||||||
| TARCA 25. |
48.7 ±12.8 |
58.4 ±12.1 | 74.1 | 211 (29.3) | 169 (23.4) | 8,855 | 22,110 | 64.0 | 81.6 | 1.5 | 567 (8.9) | 1,147 (17.9) | 5.9 | |
| Anonymous submission | ||||||||||||||
| mLK 26. |
35.1 ±12.9 |
47.5 ±11.5 | 71.5 | 89 (12.3) | 276 (38.3) | 5,678 | 33,815 | 45.0 | 83.0 | 1.0 | 383 (8.5) | 1,175 (26.1) | 1.0 | |
| Yuan Zhang, Di Xie and Shiliang Pu (Hikvision Research Institute) | ||||||||||||||
| RSCNN 27. |
29.5 ±23.4 |
37.0 ±11.4 | 73.1 | 93 (12.9) | 262 (36.3) | 11,866 | 30,474 | 50.4 | 72.3 | 2.1 | 976 (19.4) | 1,176 (23.3) | 4.0 | |
| Heba Mahgoub, Khaled Mostafa, Khaled T. Wassif and Ibrahim Farag, “Multi-Target Tracking Using Hierarchical Convolutional Features and Motion Cues” International Journal of Advanced Computer Science and Applications(IJACSA), 8(11), 2017. | ||||||||||||||
| SCEA 28. |
29.1 ±12.2 |
37.2 ±10.3 | 71.1 | 64 (8.9) | 341 (47.3) | 6,060 | 36,912 | 39.9 | 80.2 | 1.0 | 604 (15.1) | 1,182 (29.6) | 6.8 | |
| J. Yoon, C. Lee, M. Yang, K. Yoon. Online Multi-object Tracking via Structural Constraint Event Aggregation. In IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2016. | ||||||||||||||
| SAS_MOT15 29. |
22.2 ±13.8 |
27.2 ±11.2 | 71.1 | 22 (3.1) | 444 (61.6) | 5,591 | 41,531 | 32.4 | 78.1 | 1.0 | 700 (21.6) | 1,240 (38.3) | 8.9 | |
| A. Maksai, P. Fua. Eliminating Exposure Bias and Metric Mismatch in Multiple Object Tracking. In CVPR, 2019. | ||||||||||||||
| OMT_DFH 30. |
21.2 ±0.0 |
37.3 ±0.0 | 69.9 | 51 (7.1) | 335 (46.5) | 13,218 | 34,657 | 43.6 | 67.0 | 2.3 | 563 (12.9) | 1,255 (28.8) | 28.6 | |
| J. Ju, D. Kim, B. Ku, D. Han, H. Ko. Online multi-object tracking with efficient track drift and fragmentation handling. In J. Opt. Soc. Am. A, 2017. | ||||||||||||||
| Tracker | MOTA | IDF1 | MOTP | MT | ML | FP | FN | Recall | Precision | FAF | ID Sw. | Frag | Hz | |
| AP_HWDPL_p 31. |
38.5 ±8.8 |
47.1 ±9.3 | 72.6 | 63 (8.7) | 270 (37.4) | 4,005 | 33,203 | 46.0 | 87.6 | 0.7 | 586 (12.8) | 1,263 (27.5) | 6.7 | |
| C. Long, A. Haizhou, S. Chong, Z. Zijie, B. Bo. Online Multi-Object Tracking with Convolutional Neural Networks. In 2017 IEEE International Conference on Image Processing (ICIP), 2017. | ||||||||||||||
| ISE_MOT15R 32. |
46.7 ±16.1 |
51.6 ±9.9 | 77.1 | 212 (29.4) | 185 (25.7) | 11,003 | 20,839 | 66.1 | 78.7 | 1.9 | 878 (13.3) | 1,265 (19.1) | 6.7 | |
| MIFT | ||||||||||||||
| GMPHD 33. |
18.5 ±12.7 |
28.4 ±8.4 | 70.9 | 28 (3.9) | 399 (55.3) | 7,864 | 41,766 | 32.0 | 71.4 | 1.4 | 459 (14.3) | 1,266 (39.5) | 19.8 | |
| Y. Song, M. Jeon. Online Multiple Object Tracking with the Hierarchically Adopted GM-PHD Filter using Motion and Appearance. In IEEE/IEIE The International Conference on Consumer Electronics (ICCE) Asia, 2016. | ||||||||||||||
| HybridDAT 34. |
35.0 ±13.5 |
47.7 ±9.6 | 72.6 | 82 (11.4) | 304 (42.2) | 8,455 | 31,140 | 49.3 | 78.2 | 1.5 | 358 (7.3) | 1,267 (25.7) | 4.6 | |
| M. Yang, Y. Jia. A Hybrid Data Association Framework for Robust Online Multi-Object Tracking. In IEEE Transactions on Image Processing, 2016. | ||||||||||||||
| RMOT 35. |
18.6 ±0.0 |
32.6 ±0.0 | 69.6 | 38 (5.3) | 384 (53.3) | 12,473 | 36,835 | 40.0 | 66.4 | 2.2 | 684 (17.1) | 1,282 (32.0) | 7.9 | |
| J. Yoon, H. Yang, J. Lim, K. Yoon. Bayesian Multi-Object Tracking Using Motion Context from Multiple Objects. In IEEE Winter Conference on Applications of Computer Vision (WACV), 2015. | ||||||||||||||
| MANET 36. |
47.4 ±10.5 |
49.5 ±9.7 | 76.1 | 173 (24.0) | 193 (26.8) | 6,044 | 25,164 | 59.0 | 85.7 | 1.0 | 1,087 (18.4) | 1,290 (21.8) | 11.9 | |
| Anonymous submission | ||||||||||||||
| AdTobKF 37. |
24.8 ±12.4 |
34.5 ±11.9 | 70.8 | 29 (4.0) | 375 (52.0) | 6,201 | 39,321 | 36.0 | 78.1 | 1.1 | 666 (18.5) | 1,300 (36.1) | 206.5 | |
| K. Loumponias, A. Dimou, N. Vretos, P. Daras. Adaptive Tobit Kalman-Based Tracking. In 2018 14th International Conference on Signal-Image Technology \& Internet-Based Systems (SITIS), 2018. | ||||||||||||||
| TENSOR 38. |
24.3 ±13.1 |
24.1 ±7.8 | 71.6 | 40 (5.5) | 336 (46.6) | 6,644 | 38,582 | 37.2 | 77.5 | 1.1 | 1,271 (34.2) | 1,304 (35.1) | 24.0 | |
| X. Shi, H. Ling, Y. Pang, W. Hu, P. Chu, J. Xing. Rank-1 Tensor Approximation for High-Order Association in Multi-target Tracking. In IJCV, 2019. | ||||||||||||||
| SiameseCNN 39. |
29.0 ±15.1 |
34.3 ±14.4 | 71.2 | 61 (8.5) | 349 (48.4) | 5,160 | 37,798 | 38.5 | 82.1 | 0.9 | 639 (16.6) | 1,316 (34.2) | 52.8 | |
| Laura Leal-Taixé, Cristian Canton-Ferrer, Konrad Schindler. Learning by Tracking: Siamese CNN for Robust Target Association. DeepVision Workshop (CVPR), Las Vegas (Nevada, USA), June 2016. | ||||||||||||||
| GMPHD_OGM 40. |
30.7 ±12.6 |
38.8 ±9.3 | 71.6 | 83 (11.5) | 275 (38.1) | 6,502 | 35,030 | 43.0 | 80.2 | 1.1 | 1,034 (24.1) | 1,351 (31.4) | 169.5 | |
| Y. Song, K. Yoon, Y. Yoon, K. Yow, M. Jeon. Online Multi-Object Tracking with GMPHD Filter and Occlusion Group Management. In IEEE Access, 2019. | ||||||||||||||
| Tracker | MOTA | IDF1 | MOTP | MT | ML | FP | FN | Recall | Precision | FAF | ID Sw. | Frag | Hz | |
| TLO 41. |
41.3 ±0.0 |
46.1 ±0.0 | 73.5 | 113 (15.7) | 249 (34.5) | 8,000 | 27,210 | 55.7 | 81.1 | 1.4 | 852 (15.3) | 1,405 (25.2) | 5.0 | |
| Anonymous submission | ||||||||||||||
| QuadMOT 42. |
33.8 ±13.8 |
40.4 ±9.2 | 73.4 | 93 (12.9) | 266 (36.9) | 7,898 | 32,061 | 47.8 | 78.8 | 1.4 | 703 (14.7) | 1,430 (29.9) | 3.7 | |
| J. Son, M. Baek, M. Cho, B. Han. Multi-Object Tracking with Quadruplet Convolutional Neural Networks. In CVPR, 2017. | ||||||||||||||
| KCF 43. |
38.9 ±13.9 |
44.5 ±11.2 | 70.6 | 120 (16.6) | 227 (31.5) | 7,321 | 29,501 | 52.0 | 81.4 | 1.3 | 720 (13.9) | 1,440 (27.7) | 0.3 | |
| P. Chu, H. Fan, C. Tan, H. Ling. Online Multi-Object Tracking with Instance-Aware Tracker and Dynamic Model Refreshment. In WACV, 2019. | ||||||||||||||
| AM 44. |
34.3 ±13.1 |
48.3 ±12.2 | 70.5 | 82 (11.4) | 313 (43.4) | 5,154 | 34,848 | 43.3 | 83.8 | 0.9 | 348 (8.0) | 1,463 (33.8) | 0.5 | |
| Q. Chu, W. Ouyang, H. Li, X. Wang, B. Liu, N. Yu. Online Multi-object Tracking Using CNN-Based Single Object Tracker with Spatial-Temporal Attention Mechanism. In 2017 IEEE International Conference on Computer Vision (ICCV), 2017. | ||||||||||||||
| EAMTTpub 45. |
22.3 ±14.2 |
32.8 ±13.0 | 70.8 | 39 (5.4) | 380 (52.7) | 7,924 | 38,982 | 36.6 | 73.9 | 1.4 | 833 (22.8) | 1,485 (40.6) | 12.2 | |
| R. Sanchez-Matilla, F. Poiesi, A. Cavallaro "Multi-target tracking with strong and weak detections" in BMTT ECCVw 2016 | ||||||||||||||
| MDP 46. |
30.3 ±0.0 |
44.7 ±0.0 | 71.3 | 94 (13.0) | 277 (38.4) | 9,717 | 32,422 | 47.2 | 74.9 | 1.7 | 680 (14.4) | 1,500 (31.8) | 1.1 | |
| Y. Xiang, A. Alahi, S. Savarese. Learning to Track: Online Multi-Object Tracking by Decision Making. In International Conference on Computer Vision (ICCV), 2015. | ||||||||||||||
| TDAM 47. |
33.0 ±0.0 |
46.1 ±0.0 | 72.8 | 96 (13.3) | 282 (39.1) | 10,064 | 30,617 | 50.2 | 75.4 | 1.7 | 464 (9.2) | 1,506 (30.0) | 5.9 | |
| M. Yang, Y. Jia. Temporal dynamic appearance modeling for online multi-person tracking. In Computer Vision and Image Understanding, 2016. | ||||||||||||||
| CRFTrack_ 48. |
40.0 ±14.5 |
49.6 ±14.7 | 71.9 | 166 (23.0) | 206 (28.6) | 10,295 | 25,917 | 57.8 | 77.5 | 1.8 | 658 (11.4) | 1,508 (26.1) | 3.2 | |
| Jun xiang, Chao Ma, Guohan Xu, Jianhua Hou, End-to-End Learning Deep CRF models for Multi-Object Tracking. In IEEE Transactions on Circuits and Systems for Video Technology, 2020 | ||||||||||||||
| TBSS15 49. |
29.2 ±12.4 |
37.2 ±9.0 | 71.3 | 49 (6.8) | 316 (43.8) | 6,068 | 36,779 | 40.1 | 80.3 | 1.1 | 649 (16.2) | 1,508 (37.6) | 11.5 | |
| X. Zhou, P. Jiang, Z. Wei, H. Dong, F. Wang. Online Multi-Object Tracking with Structural Invariance Constraint. In BMVC, 2018. | ||||||||||||||
| RAR15pub 50. |
35.1 ±12.4 |
45.4 ±11.4 | 70.9 | 94 (13.0) | 305 (42.3) | 6,771 | 32,717 | 46.7 | 80.9 | 1.2 | 381 (8.1) | 1,523 (32.6) | 5.4 | |
| K. Fang, Y. Xiang, X. Li, S. Savarese. Recurrent Autoregressive Networks for Online Multi-Object Tracking. In The IEEE Winter Conference on Applications of Computer Vision (WACV), 2018. | ||||||||||||||
| Tracker | MOTA | IDF1 | MOTP | MT | ML | FP | FN | Recall | Precision | FAF | ID Sw. | Frag | Hz | |
| TBX 51. |
27.5 ±13.3 |
33.8 ±12.1 | 70.6 | 75 (10.4) | 330 (45.8) | 7,968 | 35,810 | 41.7 | 76.3 | 1.4 | 759 (18.2) | 1,528 (36.6) | 0.1 | |
| R. Henschel, L. Leal-Taixé, B. Rosenhahn, K. Schindler. Tracking with multi-level features. In arXiv:1607.07304, 2016. | ||||||||||||||
| DEEPDA_MOT 52. |
22.5 ±0.0 |
25.9 ±0.0 | 70.9 | 46 (6.4) | 447 (62.0) | 7,346 | 39,092 | 36.4 | 75.3 | 1.3 | 1,159 (31.9) | 1,538 (42.3) | 172.8 | |
| K. Yoon, D. Kim, Y. Yoon, M. Jeon. Data Association for Multi-Object Tracking via Deep Neural Networks. In Sensors, 2019. | ||||||||||||||
| TSDA_OAL 53. |
18.6 ±0.0 |
36.1 ±0.0 | 69.7 | 68 (9.4) | 305 (42.3) | 16,350 | 32,853 | 46.5 | 63.6 | 2.8 | 806 (17.3) | 1,544 (33.2) | 19.7 | |
| H. Ko. Online multi-person tracking with two-stage data association and online appearance model learning. In IET Computer Vision, 2017. | ||||||||||||||
| DCCRF 54. |
33.6 ±10.9 |
39.1 ±10.1 | 70.9 | 75 (10.4) | 271 (37.6) | 5,917 | 34,002 | 44.7 | 82.3 | 1.0 | 866 (19.4) | 1,566 (35.1) | 0.1 | |
| H. Zhou, W. Ouyang, J. Cheng, X. Wang, H. Li. Deep Continuous Conditional Random Fields with Asymmetric Inter-object Constraints for Online Multi-object Tracking. In IEEE Transactions on Circuits and Systems for Video Technology, 2018. | ||||||||||||||
| CDA_DDALpb 55. |
32.8 ±10.6 |
38.8 ±8.0 | 70.7 | 70 (9.7) | 304 (42.2) | 4,983 | 35,690 | 41.9 | 83.8 | 0.9 | 614 (14.7) | 1,583 (37.8) | 2.3 | |
| S. Bae and K. Yoon, Confidence-Based Data Association and Discriminative Deep Appearance Learning for Robust Online Multi-Object Tracking , In IEEE TPAMI, 2017. | ||||||||||||||
| TLO15 56. |
40.0 ±14.9 |
44.3 ±9.9 | 73.4 | 123 (17.1) | 208 (28.8) | 9,349 | 26,328 | 57.1 | 79.0 | 1.6 | 1,207 (21.1) | 1,624 (28.4) | 24.6 | |
| Anonymous submission | ||||||||||||||
| oICF 57. |
27.1 ±14.9 |
40.5 ±12.6 | 70.0 | 46 (6.4) | 351 (48.7) | 7,594 | 36,757 | 40.2 | 76.5 | 1.3 | 454 (11.3) | 1,660 (41.3) | 1.4 | |
| H. Kieritz, S. Becker, W. Hübner, M. Arens. Online Multi-Person Tracking using Integral Channel Features. In IEEE Advanced Video and Signal-based Surveillance (AVSS) 2016, 2016. | ||||||||||||||
| Tracktor++v2 58. |
46.6 ±10.0 |
47.6 ±10.2 | 76.4 | 131 (18.2) | 201 (27.9) | 4,624 | 26,896 | 56.2 | 88.2 | 0.8 | 1,290 (22.9) | 1,702 (30.3) | 1.4 | |
| P. Bergmann, T. Meinhardt, L. Leal-Taixé. Tracking without bells and whistles. In ICCV, 2019. | ||||||||||||||
| LP2D 59. |
19.8 ±37.1 |
0.0 ±0.0 | 71.2 | 48 (6.7) | 297 (41.2) | 11,580 | 36,045 | 41.3 | 68.7 | 2.0 | 1,649 (39.9) | 1,712 (41.4) | inf | |
| MOT baseline: Linear programming on 2D image coordinates. | ||||||||||||||
| TC_ODAL 60. |
15.1 ±37.1 |
0.0 ±0.0 | 70.5 | 23 (3.2) | 402 (55.8) | 12,970 | 38,538 | 37.3 | 63.8 | 2.2 | 637 (17.1) | 1,716 (46.0) | 1.5 | |
| S. Bae, K. Yoon. Robust Online Multi-Object Tracking based on Tracklet Confidence and Online Discriminative Appearance Learning. In CVPR, 2014. | ||||||||||||||
| Tracker | MOTA | IDF1 | MOTP | MT | ML | FP | FN | Recall | Precision | FAF | ID Sw. | Frag | Hz | |
| Tracktor++ 61. |
44.1 ±11.7 |
46.7 ±10.3 | 75.0 | 130 (18.0) | 189 (26.2) | 6,477 | 26,577 | 56.7 | 84.3 | 1.1 | 1,318 (23.2) | 1,790 (31.5) | 0.9 | |
| P. Bergmann, T. Meinhardt, L. Leal-Taixé. Tracking without bells and whistles. In ICCV, 2019. | ||||||||||||||
| ELP 62. |
25.0 ±0.0 |
26.2 ±0.0 | 71.2 | 54 (7.5) | 316 (43.8) | 7,345 | 37,344 | 39.2 | 76.6 | 1.3 | 1,396 (35.6) | 1,804 (46.0) | 5.7 | |
| N. McLaughlin, J. Martinez Del Rincon, P. Miller. Enhancing Linear Programming with Motion Modeling for Multi-target Tracking. In IEEE Winter Conference on Applications of Computer Vision (WACV), 2015. | ||||||||||||||
| SC2D 63. |
50.1 ±19.0 |
55.7 ±7.7 | 77.2 | 349 (48.4) | 74 (10.3) | 15,312 | 14,531 | 76.3 | 75.4 | 2.7 | 797 (10.4) | 1,816 (23.8) | 22.6 | |
| Anonymous submission | ||||||||||||||
| RAT 64. |
19.0 ±10.6 |
26.1 ±7.7 | 70.2 | 16 (2.2) | 369 (51.2) | 7,534 | 40,858 | 33.5 | 73.2 | 1.3 | 1,401 (41.8) | 1,826 (54.5) | 21.8 | |
| Anonymous submission | ||||||||||||||
| GMMA_intp 65. |
27.3 ±12.1 |
36.6 ±8.5 | 70.9 | 47 (6.5) | 311 (43.1) | 7,848 | 35,817 | 41.7 | 76.6 | 1.4 | 987 (23.7) | 1,848 (44.3) | 132.5 | |
| Y. Song, Y. Yoon, K. Yoon, M. Jeon. Online and Real-Time Tracking with the GMPHD Filter using Group Management and Relative Motion Analysis. In Proc. IEEE Int. Workshop Traffic Street Surveill. Safety Secur. (AVSS), 2018. | ||||||||||||||
| EDA_GNN 66. |
21.8 ±13.8 |
27.8 ±5.8 | 70.5 | 65 (9.0) | 290 (40.2) | 11,970 | 34,587 | 43.7 | 69.2 | 2.1 | 1,488 (34.0) | 1,851 (42.4) | 56.4 | |
| Paper ID 2713 | ||||||||||||||
| TrctrD15 67. |
44.1 ±11.7 |
46.0 ±10.6 | 75.3 | 124 (17.2) | 192 (26.6) | 6,085 | 26,917 | 56.2 | 85.0 | 1.1 | 1,347 (24.0) | 1,868 (33.2) | 1.6 | |
| Y. Xu, A. Osep, Y. Ban, R. Horaud, L. Leal-Taixe, X. Alameda-Pineda. How To Train Your Deep Multi-Object Tracker. In , 2019. | ||||||||||||||
| ALExTRAC 68. |
17.0 ±12.1 |
17.3 ±8.8 | 71.2 | 28 (3.9) | 378 (52.4) | 9,233 | 39,933 | 35.0 | 70.0 | 1.6 | 1,859 (53.1) | 1,872 (53.5) | 3.7 | |
| A. Bewley, L. Ott, F. Ramos, B. Upcroft. ALExTRAC: Affinity Learning by Exploring Temporal Reinforcement within Association Chains. In International Conference on Robotics and Automation (ICRA), (to appear) 2016. | ||||||||||||||
| TBD 69. |
15.9 ±37.1 |
0.0 ±0.0 | 70.9 | 46 (6.4) | 345 (47.9) | 14,943 | 34,777 | 43.4 | 64.1 | 2.6 | 1,939 (44.7) | 1,963 (45.2) | inf | |
| A. Geiger, M. Lauer, C. Wojek, C. Stiller, R. Urtasun. 3D Traffic Scene Understanding from Movable Platforms. In Pattern Analysis and Machine Intelligence (PAMI), 2014. | ||||||||||||||
| CppSORT 70. |
21.7 ±11.8 |
26.8 ±9.3 | 71.2 | 27 (3.7) | 354 (49.1) | 8,422 | 38,454 | 37.4 | 73.2 | 1.5 | 1,231 (32.9) | 2,005 (53.6) | 1,112.1 | |
| S. Murray. Real-Time Multiple Object Tracking - A Study on the Importance of Speed. In arXiv preprint arXiv:1709.03572, 2017. | ||||||||||||||
| Tracker | MOTA | IDF1 | MOTP | MT | ML | FP | FN | Recall | Precision | FAF | ID Sw. | Frag | Hz | |
| AMIR15 71. |
37.6 ±12.6 |
46.0 ±7.9 | 71.7 | 114 (15.8) | 193 (26.8) | 7,933 | 29,397 | 52.2 | 80.2 | 1.4 | 1,026 (19.7) | 2,024 (38.8) | 1.9 | |
| A. Sadeghian, A. Alahi, S. Savarese. Tracking The Untrackable: Learning To Track Multiple Cues with Long-Term Dependencies. In ICCV, 2017. | ||||||||||||||
| RNN_LSTM 72. |
19.0 ±18.1 |
17.1 ±8.3 | 71.0 | 40 (5.5) | 329 (45.6) | 11,578 | 36,706 | 40.3 | 68.1 | 2.0 | 1,490 (37.0) | 2,081 (51.7) | 165.2 | |
| A. Milan, S. Rezatofighi, A. Dick, I. Reid, K. Schindler. Online Multi-Target Tracking using Recurrent Neural Networks. In AAAI, 2017. | ||||||||||||||
| SMOT 73. |
18.2 ±37.1 |
0.0 ±0.0 | 71.2 | 20 (2.8) | 395 (54.8) | 8,780 | 40,310 | 34.4 | 70.6 | 1.5 | 1,148 (33.4) | 2,132 (62.0) | 2.7 | |
| C. Dicle, O. Camps, M. Sznaier. The Way They Move: Tracking Targets with Similar Appearance. In ICCV, 2013. | ||||||||||||||
| PHD_GSDL 74. |
30.5 ±14.2 |
38.8 ±10.0 | 71.2 | 55 (7.6) | 297 (41.2) | 6,534 | 35,284 | 42.6 | 80.0 | 1.1 | 879 (20.6) | 2,208 (51.9) | 8.2 | |
| Z. Fu, P. Feng, F. Angelini, J. Chambers, S. Naqvi. Particle PHD Filter based Multiple Human Tracking using Online Group-Structured Dictionary Learning. In IEEE Access, 2018. | ||||||||||||||
| MTSTracker 75. |
20.6 ±18.2 |
31.9 ±20.6 | 70.3 | 65 (9.0) | 266 (36.9) | 15,161 | 32,212 | 47.6 | 65.8 | 2.6 | 1,387 (29.2) | 2,357 (49.5) | 19.3 | |
| F. Nguyen Thi Lan Anh, F. Bremond. Multi-Object Tracking using Multi-Channel Part Appearance Representation. In International conference on Advanced video and Signal Based Surveillance, 2017. | ||||||||||||||
| Tracker 76. |
16.8 ±11.3 |
22.2 ±6.1 | 70.1 | 15 (2.1) | 389 (54.0) | 8,519 | 40,697 | 33.8 | 70.9 | 1.5 | 1,917 (56.8) | 2,416 (71.6) | 9.5 | |
| Anonymous submission | ||||||||||||||
| STRN 77. |
38.1 ±11.3 |
46.6 ±7.8 | 72.1 | 83 (11.5) | 241 (33.4) | 5,451 | 31,571 | 48.6 | 84.6 | 0.9 | 1,033 (21.2) | 2,665 (54.8) | 13.8 | |
| J. Xu, Y. Cao, Z. Zhang, H. Hu. Spatial-Temporal Relation Networks for Multi-Object Tracking. In ICCV, 2019. | ||||||||||||||
| INARLA 78. |
34.7 ±13.2 |
42.1 ±9.7 | 70.7 | 90 (12.5) | 216 (30.0) | 9,855 | 29,158 | 52.5 | 76.6 | 1.7 | 1,112 (21.2) | 2,848 (54.2) | 2.6 | |
| H. Wu, Y. Hu, K. Wang, H. Li, L. Nie, H. Cheng. Instance-aware representation learning and association for online multi-person tracking. In Pattern Recognition, 2019. | ||||||||||||||
| LDCT 79. |
4.7 ±41.3 |
16.8 ±18.8 | 71.7 | 82 (11.4) | 234 (32.5) | 14,066 | 32,156 | 47.7 | 67.6 | 2.4 | 12,348 (259.1) | 2,918 (61.2) | 20.7 | |
| F. Solera, S. Calderara, R. Cucchiara. Learning to Divide and Conquer for Online Multi-Target Tracking. In ICCV, 2015 | ||||||||||||||
| DP_NMS 80. |
14.5 ±15.2 |
19.7 ±6.5 | 70.8 | 43 (6.0) | 294 (40.8) | 13,171 | 34,814 | 43.3 | 66.9 | 2.3 | 4,537 (104.7) | 3,090 (71.3) | 444.8 | |
| H. Pirsiavash, D. Ramanan, C. Fowlkes. Globally-Optimal Greedy Algorithms for Tracking a Variable Number of Objects. In CVPR, 2011. | ||||||||||||||
| Sequences | Frames | Trajectories | Boxes |
| 11 | 5783 | 721 | 61440 |
Difficulty Analysis
Sequence difficulty (from easiest to hardest, measured by average MOTA)
TUD-Crossing
(64.3 MOTA)
PETS09-S2L2
(43.6 MOTA)
ETH-Jelmoli
(40.6 MOTA)
...
...
Venice-1
(24.0 MOTA)
ADL-Rundle-1
(18.0 MOTA)
Evaluation Measures
Lower is better. Higher is better.| Measure | Better | Perfect | Description |
| MOTA | higher | 100 % | Multiple Object Tracking Accuracy [1]. This measure combines three error sources: false positives, missed targets and identity switches. |
| MOTP | higher | 100 % | Multiple Object Tracking Precision [1]. The misalignment between the annotated and the predicted bounding boxes. |
| IDF1 | higher | 100 % | ID F1 Score [2]. The ratio of correctly identified detections over the average number of ground-truth and computed detections. |
| FAF | lower | 0 | The average number of false alarms per frame. |
| MT | higher | 100 % | Mostly tracked targets. The ratio of ground-truth trajectories that are covered by a track hypothesis for at least 80% of their respective life span. |
| ML | lower | 0 % | Mostly lost targets. The ratio of ground-truth trajectories that are covered by a track hypothesis for at most 20% of their respective life span. |
| FP | lower | 0 | The total number of false positives. |
| FN | lower | 0 | The total number of false negatives (missed targets). |
| ID Sw. | lower | 0 | The total number of identity switches. Please note that we follow the stricter definition of identity switches as described in [3]. |
| Frag | lower | 0 | The total number of times a trajectory is fragmented (i.e. interrupted during tracking). |
| Hz | higher | Inf. | Processing speed (in frames per second excluding the detector) on the benchmark. The frequency is provided by the authors and not officially evaluated by the MOTChallenge. |
Legend
| Symbol | Description |
|
This is an online (causal) method, i.e. the solution is immediately available with each incoming frame and cannot be changed at any later time. |
|
This method used the provided detection set as input. |
 |
This method used the provided detection set as input. |
 |
This entry has been submitted or updated less than a week ago. |
|
This entry has been submitted or updated less than a week ago. |
References:
| [1] | Evaluating Multiple Object Tracking Performance: The CLEAR MOT Metrics. Image and Video Processing, 2008(1):1-10, 2008. |
| [2] | Performance Measures and a Data Set for Multi-Target, Multi-Camera Tracking. In ECCV workshop on Benchmarking Multi-Target Tracking, 2016. |
| [3] | Learning to associate: HybridBoosted multi-target tracker for crowded scene. In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2009. |
