MOT16 Results

Click on a measure to sort the table accordingly. See below for a more detailed description.


Showing only entries that use public detections!


Due to a minor bug in the export script, all results were re-evaluated on April 11, 2016. Here is the old snapshot of the leaderboard.


Benchmark Statistics

TrackerMOTAIDF1MOTPMTMLFPFNRecallPrecisionFAFID Sw.FragHz
AFN
1. using public detections
49.0
±10.2
48.2
±7.4
78.0 145 (19.1)271 (35.7)9,508 82,506 54.7 91.3 1.6 899 (16.4)1,383 (25.3)0.6
H. Shen, L. Huang, C. Huang, W. Xu. Tracklet Association Tracker: An End-to-End Learning-based Association Approach for Multi-Object Tracking. In CoRR, 2018.
AMIR
2. online method using public detections
47.2
±8.2
46.3
±7.2
75.8 106 (14.0)316 (41.6)2,681 92,856 49.1 97.1 0.5 774 (15.8)1,675 (34.1)1.0
A. Sadeghian, A. Alahi, S. Savarese. Tracking The Untrackable: Learning To Track Multiple Cues with Long-Term Dependencies. In ICCV, 2017.
AM_ADM
3. online method using public detections
40.1
±10.1
43.8
±9.6
75.4 54 (7.1)351 (46.2)8,503 99,891 45.2 90.6 1.4 789 (17.5)1,736 (38.4)5.8
S. Lee, M. Kim, S. Bae, Learning Discriminative Appearance Models for Online Multi-Object Tracking with Appearance Discriminability Measures, In IEEE Access, 2018.
ASTT
4. using public detections
47.2
±9.6
44.3
±8.5
76.1 124 (16.3)316 (41.6)4,680 90,877 50.2 95.1 0.8 633 (12.6)814 (16.2)0.5
Yi Tao el al., “Adaptive Spatio-temporal Model Based Multiple Object Tracking Considering a Moving Camera[C]”, International Conference on Universal Village (UV), 2018.
CDA_DDALv2
5. online method using public detections
43.9
±7.8
45.1
±5.7
74.7 81 (10.7)337 (44.4)6,450 95,175 47.8 93.1 1.1 676 (14.1)1,795 (37.6)0.5
S. Bae and K. Yoon, Confidence-Based Data Association and Discriminative Deep Appearance Learning for Robust Online Multi-Object Tracking , In IEEE TPAMI, 2017.
CEM
6. using public detections
33.2
±8.8
0.0
±0.0
75.8 59 (7.8)413 (54.4)6,837 114,322 37.3 90.9 1.2 642 (17.2)731 (19.6)5,919.0
A. Milan, S. Roth, K. Schindler. Continuous Energy Minimization for Multitarget Tracking. In IEEE TPAMI, 2014.
CMT16
7. using public detections
49.8
±9.0
59.2
±6.5
76.1 126 (16.6)331 (43.6)9,229 81,882 55.1 91.6 1.6 365 (6.6)617 (11.2)6.3
#Submission: TIP-21190-2019
CppSORT
8. online method using public detections
31.5
±9.0
27.7
±9.7
77.3 33 (4.3)455 (59.9)3,048 120,278 34.0 95.3 0.5 1,587 (46.6)2,239 (65.8)687.1
S. Murray. Real-Time Multiple Object Tracking - A Study on the Importance of Speed. In arXiv preprint arXiv:1709.03572, 2017.
CRF_TRACK
9. using public detections
50.3
±7.9
54.4
±6.4
74.8 139 (18.3)271 (35.7)7,148 82,746 54.6 93.3 1.2 702 (12.9)1,387 (25.4)1.5
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
DASOT16
10. online method using public detections
46.1
±9.2
49.4
±8.8
75.3 111 (14.6)316 (41.6)8,222 89,204 51.1 91.9 1.4 802 (15.7)2,057 (40.3)9.0
Q. Chu, W. Ouyang, B. Liu, F. Zhu, N. Yu. DASOT: A Unified Framework Integrating Data Association and Single Object Tracking for Online Multi-Object Tracking. In Proceedings of the AAAI Conference on Artificial Intelligence, 2020.
TrackerMOTAIDF1MOTPMTMLFPFNRecallPrecisionFAFID Sw.FragHz
DAST
11. online method using public detections
48.9
±8.5
53.2
±5.5
74.9 115 (15.2)275 (36.2)9,987 82,427 54.8 90.9 1.7 838 (15.3)1,936 (35.3)8.7
Anonymous submission
DCCRF16
12. online method using public detections
44.8
±9.5
39.7
±8.2
75.6 107 (14.1)321 (42.3)5,613 94,133 48.4 94.0 0.9 968 (20.0)1,378 (28.5)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.
DD_TAMA16
13. online method using public detections
46.2
±8.4
49.4
±7.6
75.4 107 (14.1)334 (44.0)5,126 92,367 49.3 94.6 0.9 598 (12.1)1,127 (22.8)6.5
Y. Yoon, D. Kim, K. Yoon, Y. Song, M. Jeon. Online Multiple Pedestrian Tracking using Deep Temporal Appearance Matching Association. In arXiv:1907.00831, 2019.
deepS2
14. using public detections
46.0
±8.2
46.5
±6.1
76.3 118 (15.5)323 (42.6)5,124 92,697 49.2 94.6 0.9 693 (14.1)759 (15.4)0.7
ID 32
DMAN
15. online method using public detections
46.1
±9.1
54.8
±7.0
73.8 132 (17.4)324 (42.7)7,909 89,874 50.7 92.1 1.3 532 (10.5)1,616 (31.9)0.3
J. Zhu, H. Yang, N. Liu, M. Kim, W. Zhang, M. Yang. Online Multi-Object Tracking with Dual Matching Attention Networks. In ECCV, 2018.
DN_MOT
16. online method using public detections
58.8
±9.3
60.8
±7.7
74.8 233 (30.7)139 (18.3)18,208 55,366 69.6 87.5 3.1 1,626 (23.4)2,904 (41.7)22.3
Anonymous submission
DP_NMS
17. using public detections
26.2
±9.7
31.2
±5.1
76.3 31 (4.1)512 (67.5)3,689 130,557 28.4 93.3 0.6 365 (12.9)638 (22.5)212.6
H. Pirsiavash, D. Ramanan, C. Fowlkes. Globally-Optimal Greedy Algorithms for Tracking a Variable Number of Objects. In CVPR, 2011.
EAGS16
18. using public detections
47.4
±8.6
50.1
±7.4
75.9 131 (17.3)324 (42.7)8,369 86,931 52.3 91.9 1.4 575 (11.0)913 (17.5)197.3
H. Sheng, X. Zhang, Y. Zhang, Y. Wu, J. Chen. Enhanced Association with Supervoxels in Multiple Hypothesis Tracking. In IEEE Access, 2018.
EAMTT_pub
19. online method using public detections
38.8
±8.6
42.4
±7.4
75.1 60 (7.9)373 (49.1)8,114 102,452 43.8 90.8 1.4 965 (22.0)1,657 (37.8)11.8
R. Sanchez-Matilla, F. Poiesi, A. Cavallaro "Multi-target tracking with strong and weak detections" in BMTT ECCVw 2016
EDMT
20. using public detections
45.3
±9.1
47.9
±7.8
75.9 129 (17.0)303 (39.9)11,122 87,890 51.8 89.5 1.9 639 (12.3)946 (18.3)1.8
J. Chen, H. Sheng, Y. Zhang, Z. Xiong. Enhancing Detection Model for Multiple Hypothesis Tracking. In BMTT-PETS CVPRw, 2017.
TrackerMOTAIDF1MOTPMTMLFPFNRecallPrecisionFAFID Sw.FragHz
eHAF16
21. using public detections
47.2
±8.7
52.4
±7.6
75.7 141 (18.6)325 (42.8)12,586 83,107 54.4 88.7 2.1 542 (10.0)787 (14.5)0.5
H. Sheng, Y. Zhang, J. Chen, Z. Xiong, J. Zhang. Heterogeneous Association Graph Fusion for Target Association in Multiple Object Tracking. In IEEE Transactions on Circuits and Systems for Video Technology, 2018.
ENFT16
22. using public detections
50.3
±8.2
55.0
±6.4
76.2 146 (19.2)302 (39.8)8,341 81,843 55.1 92.3 1.4 490 (8.9)754 (13.7)0.4
BUAA
eTC
23. using public detections
49.2
±9.1
56.1
±7.2
75.5 131 (17.3)306 (40.3)8,400 83,702 54.1 92.2 1.4 606 (11.2)882 (16.3)0.7
G. Wang, Y. Wang, H. Zhang, R. Gu, J. Hwang. Exploit the connectivity: Multi-object tracking with trackletnet. In Proceedings of the 27th ACM International Conference on Multimedia, 2019.
FWT
24. using public detections
47.8
±10.0
44.3
±11.0
75.5 145 (19.1)290 (38.2)8,886 85,487 53.1 91.6 1.5 852 (16.0)1,534 (28.9)0.6
R. Henschel, L. Leal-Taixé, D. Cremers, B. Rosenhahn. Fusion of Head and Full-Body Detectors for Multi-Object Tracking. In Trajnet CVPRW, 2018.
GCRA
25. using public detections
48.2
±8.3
48.6
±5.6
77.5 98 (12.9)312 (41.1)5,104 88,586 51.4 94.8 0.9 821 (16.0)1,117 (21.7)2.8
C. Ma, C. Yang, F. Yang, Y. Zhuang, Z. Zhang, H. Jia, X. Xie. Trajectory Factory: Tracklet Cleaving and Re-connection by Deep Siamese Bi-GRU for Multiple Object Tracking. In ICME, 2018.
GMMCP
26. using public detections
38.1
±7.8
35.5
±4.5
75.8 65 (8.6)386 (50.9)6,607 105,315 42.2 92.1 1.1 937 (22.2)1,669 (39.5)0.5
A. Dehghan, S. Assari, M. Shah.. GMMCP-Tracker:Globally Optimal Generalized Maximum Multi Clique Problem for Multiple Object Tracking. In CVPR, 2015.
GMPHD_HDA
27. online method using public detections
30.5
±6.9
33.4
±5.4
75.4 35 (4.6)453 (59.7)5,169 120,970 33.6 92.2 0.9 539 (16.0)731 (21.7)13.6
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.
GMPHD_ReId
28. online method using public detections
40.4
±9.3
49.7
±8.5
75.2 85 (11.2)329 (43.3)6,572 101,266 44.5 92.5 1.1 792 (17.8)2,529 (56.9)31.6
N. Baisa. Occlusion-robust Online Multi-object Visual Tracking using a GM-PHD Filter with a CNN-based Re-identification. In , 2019.
GM_PHD_DAL
29. online method using public detections
35.1
±9.1
26.6
±7.1
76.6 53 (7.0)390 (51.4)2,350 111,886 38.6 96.8 0.4 4,047 (104.8)5,338 (138.2)3.5
N. Baisa. Online Multi-object Visual Tracking using a GM-PHD Filter with Deep Appearance Learning. In 2019 22th International Conference on Information Fusion (FUSION), 2019.
GM_PHD_N1T
30. online method using public detections
33.3
±9.0
25.5
±8.2
76.8 42 (5.5)425 (56.0)1,750 116,452 36.1 97.4 0.3 3,499 (96.8)3,594 (99.5)9.9
N. Baisa, A. Wallace. Development of a N-type GM-PHD filter for multiple target, multiple type visual tracking. In Journal of Visual Communication and Image Representation, 2019.
TrackerMOTAIDF1MOTPMTMLFPFNRecallPrecisionFAFID Sw.FragHz
GNN_tracktor
31. online method using public detections
53.9
±11.1
53.7
±10.6
77.0 127 (16.7)299 (39.4)2,514 80,892 55.6 97.6 0.4 631 (11.3)894 (16.1)2.2
Anonymous submission
HCC
32. using public detections
49.3
±10.2
50.7
±7.4
79.0 135 (17.8)303 (39.9)5,333 86,795 52.4 94.7 0.9 391 (7.5)535 (10.2)0.8
L. Ma, S. Tang, M. Black, L. Gool. Customized Multi-Person Tracker. In Computer Vision -- ACCV 2018, 2018.
HDTR
33. using public detections
53.6
±8.7
46.6
±6.8
80.8 161 (21.2)281 (37.0)4,714 79,353 56.5 95.6 0.8 618 (10.9)833 (14.7)3.6
M. Babaee, A. Athar, G. Rigoll. Multiple People Tracking Using Hierarchical Deep Tracklet Re-identification. In arXiv preprint arXiv:1811.04091, 2018.
HISP_DAL
34. online method using public detections
37.4
±8.8
30.5
±6.8
76.3 58 (7.6)386 (50.9)3,222 108,865 40.3 95.8 0.5 2,101 (52.1)2,151 (53.4)3.3
N. Baisa. Robust Online Multi-target Visual Tracking using a HISP Filter with Discriminative Deep Appearance Learning. In CoRR, 2019.
HISP_T
35. online method using public detections
35.9
±8.7
28.9
±0.0
76.1 59 (7.8)380 (50.1)6,412 107,918 40.8 92.1 1.1 2,594 (63.6)2,298 (56.3)4.8
N. Baisa. Online Multi-target Visual Tracking using a HISP Filter. In Proceedings of the 13th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 5: VISAPP,, 2018.
INTERA_MOT
36. using public detections
45.4
±8.1
47.7
±9.0
74.4 137 (18.1)294 (38.7)13,407 85,547 53.1 87.8 2.3 600 (11.3)930 (17.5)4.3
L. Lan, X. Wang, S. Zhang, D. Tao, W. Gao, T. Huang. Interacting Tracklets for Multi-object Tracking. In IEEE Transactions on Image Processing, 2018.
ISE_MOT16
37. online method using public detections
60.1
±9.2
56.9
±7.5
77.6 198 (26.1)221 (29.1)6,964 65,044 64.3 94.4 1.2 739 (11.5)951 (14.8)6.9
MIFT
JCmin_MOT
38. online method using public detections
36.7
±9.1
36.2
±10.5
75.9 57 (7.5)413 (54.4)2,936 111,890 38.6 96.0 0.5 667 (17.3)831 (21.5)14.8
M. Abhijeet Boragule. Joint Cost Minimization for Multi-Object Tracking. In 2017 IEEE International Conference on Advanced Vide and Signale Based Surveillance, 2017.
JCSTD
39. online method using public detections
47.4
±8.3
41.1
±5.6
74.4 109 (14.4)276 (36.4)8,076 86,638 52.5 92.2 1.4 1,266 (24.1)2,697 (51.4)8.8
W. Tian, M. Lauer, L. Chen. Online Multi-Object Tracking Using Joint Domain Information in Traffic Scenarios. In IEEE Transactions on Intelligent Transportation Systems, 2019.
JMC
40. using public detections
46.3
±9.4
46.3
±9.2
75.7 118 (15.5)301 (39.7)6,373 90,914 50.1 93.5 1.1 657 (13.1)1,114 (22.2)0.8
S. Tang, B. Andres, M. Andriluka, B. Schiele. Multi-Person Tracking by Multicuts and Deep Matching. In BMTT, 2016.
TrackerMOTAIDF1MOTPMTMLFPFNRecallPrecisionFAFID Sw.FragHz
JPDA_m
41. using public detections
26.2
±8.8
0.0
±0.0
76.3 31 (4.1)512 (67.5)3,689 130,549 28.4 93.3 0.6 365 (12.9)638 (22.5)22.2
H. Rezatofighi, A. Milan, Z. Zhang, Q. Shi, A. Dick, I. Reid. Joint Probabilistic Data Association Revisited. In ICCV, 2015.
KCF16
42. online method using public detections
48.8
±9.6
47.2
±7.8
75.7 120 (15.8)289 (38.1)5,875 86,567 52.5 94.2 1.0 906 (17.3)1,116 (21.2)0.1
P. Chu, H. Fan, C. Tan, H. Ling. Online Multi-Object Tracking with Instance-Aware Tracker and Dynamic Model Refreshment. In WACV, 2019.
LFNF16
43. using public detections
43.6
±8.4
41.6
±7.8
76.6 101 (13.3)347 (45.7)6,616 95,363 47.7 92.9 1.1 836 (17.5)938 (19.7)0.6
Sheng H, Hao L, Chen J, et al. Robust Local Effective Matching Model for Multi-Target Tracking. In PCM, 2017
Lif_T
44. using public detections
61.3
±0.0
64.7
±0.0
78.3 205 (27.0)258 (34.0)4,844 65,401 64.1 96.0 0.8 389 (6.1)1,034 (16.1)0.5
Anonymous submission
LINF1
45. using public detections
41.0
±10.1
45.7
±8.5
74.8 88 (11.6)389 (51.3)7,896 99,224 45.6 91.3 1.3 430 (9.4)963 (21.1)4.2
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.
LMP
46. using public detections
48.8
±8.9
51.3
±6.8
79.0 138 (18.2)304 (40.1)6,654 86,245 52.7 93.5 1.1 481 (9.1)595 (11.3)0.5
S. Tang, M. Andriluka, B. Andres, B. Schiele. Multiple People Tracking with Lifted Multicut and Person Re-identification. In CVPR, 2017.
LM_NN
47. using public detections
31.0
±7.2
31.5
±8.3
78.4 56 (7.4)443 (58.4)2,451 122,649 32.7 96.1 0.4 678 (20.7)666 (20.3)3.0
ID NEUCOM-D-18-03230
LP2D
48. using public detections
35.7
±12.3
34.2
±9.3
75.8 66 (8.7)385 (50.7)5,084 111,163 39.0 93.3 0.9 915 (23.4)1,264 (32.4)49.3
MOT baseline: Linear programming on 2D image coordinates.
LSST16O
49. online method using public detections
49.2
±10.2
56.5
±7.2
74.0 102 (13.4)314 (41.4)7,187 84,875 53.4 93.1 1.2 606 (11.3)2,497 (46.7)2.0
Multi-Object Tracking with Multiple Cues and Switcher-Aware Classification
LTTSC-CRF
50. using public detections
37.6
±8.0
42.1
±6.6
75.9 73 (9.6)419 (55.2)11,969 101,343 44.4 87.1 2.0 481 (10.8)1,012 (22.8)0.6
N. Le, A. Heili, M. Odobez. Long-Term Time-Sensitive Costs for CRF-Based Tracking by Detection. In ECCVw, 2016.
TrackerMOTAIDF1MOTPMTMLFPFNRecallPrecisionFAFID Sw.FragHz
MCjoint
51. using public detections
47.1
±10.3
52.3
±10.2
76.3 155 (20.4)356 (46.9)6,703 89,368 51.0 93.3 1.1 370 (7.3)598 (11.7)0.6
}@article{DBLP:journals/corr/KeuperTYABS16, author = {Margret Keuper and Siyu Tang and Zhongjie Yu and Bjoern Andres and Thomas Brox and Bernt Schiele}, title = {A Multi-cut Formulation for Joint Segmentation and Tracking of Multiple Objects}, journal = {CoRR}, volume = {abs/1607.06317}, year = {2016}, url = {http://arxiv.org/abs/1607.06317}, timestamp = {Wed, 07 Jun 2017 14:41:31 +0200}, biburl = {http://dblp.uni-trier.de/rec/bib/journals/corr/KeuperTYABS16}, bibsource = {dblp computer science bibliography, http://dblp.org} }
MHT_bLSTM6
52. using public detections
42.1
±8.6
47.8
±7.4
75.9 113 (14.9)337 (44.4)11,637 93,172 48.9 88.5 2.0 753 (15.4)1,156 (23.6)1.8
C. Kim, F. Li, J. Rehg. Multi-object Tracking with Neural Gating Using Bilinear LSTM. In ECCV, 2018.
MHT_DAM
53. using public detections
45.8
±8.6
46.1
±7.6
76.3 123 (16.2)328 (43.2)6,412 91,758 49.7 93.4 1.1 590 (11.9)781 (15.7)0.8
C. Kim, F. Li, A. Ciptadi, J. Rehg. Multiple Hypothesis Tracking Revisited. In ICCV, 2015.
MLT
54. online method using public detections
52.8
±8.2
62.6
±6.8
76.1 160 (21.1)322 (42.4)5,362 80,444 55.9 95.0 0.9 299 (5.4)702 (12.6)5.9
Y. Zhang, H. Sheng, Y. Wu, S. Wang, W. Ke, Z. Xiong. Multiplex Labeling Graph for Near Online Tracking in Crowded Scenes. In IEEE Internet of Things Journal, 2020.
MOTDT
55. online method using public detections
47.6
±8.4
50.9
±5.5
74.8 115 (15.2)291 (38.3)9,253 85,431 53.1 91.3 1.6 792 (14.9)1,858 (35.0)20.6
C. Long, A. Haizhou, Z. Zijie, S. Chong. Real-time Multiple People Tracking with Deeply Learned Candidate Selection and Person Re-identification. In ICME, 2018.
MPNTrack
56. using public detections
58.6
±10.3
61.7
±7.3
78.9 207 (27.3)258 (34.0)4,949 70,252 61.5 95.8 0.8 354 (5.8)684 (11.1)6.5
G. Brasó, L. Leal-Taixé. Learning a Neural Solver for Multiple Object Tracking. In CVPR, 2020.
msot
57. online method using public detections
59.5
±9.0
52.0
±8.1
78.1 187 (24.6)224 (29.5)5,070 68,001 62.7 95.8 0.9 800 (12.8)1,014 (16.2)1.1
Anonymous submission
MTDF
58. online method using public detections
45.7
±10.8
40.1
±8.7
72.6 107 (14.1)276 (36.4)12,018 84,970 53.4 89.0 2.0 1,987 (37.2)3,377 (63.2)1.5
Z. Fu, F. Angelini, J. Chambers, S. Naqvi. Multi-Level Cooperative Fusion of GM-PHD Filters for Online Multiple Human Tracking. In IEEE Transactions on Multimedia, 2019.
M_track
59. online method using public detections
56.4
±11.9
54.6
±11.4
73.5 168 (22.1)248 (32.7)4,694 73,593 59.6 95.9 0.8 1,139 (19.1)1,411 (23.7)4.7
Anonymous submission
M_track
60. online method using public detections
57.8
±11.7
57.5
±10.3
78.6 197 (26.0)238 (31.4)4,193 72,224 60.4 96.3 0.7 561 (9.3)844 (14.0)4.4
Anonymous submission
TrackerMOTAIDF1MOTPMTMLFPFNRecallPrecisionFAFID Sw.FragHz
NLLMPa
61. using public detections
47.6
±10.6
47.3
±9.6
78.5 129 (17.0)307 (40.4)5,844 89,093 51.1 94.1 1.0 629 (12.3)768 (15.0)8.3
E. Levinkov, J. Uhrig, S. Tang, M. Omran, E. Insafutdinov, A. Kirillov, C. Rother, T. Brox, B. Schiele, B. Andres. Joint Graph Decomposition and Node Labeling: Problem, Algorithms, Applications. In CVPR, 2017.
NOMT
62. using public detections
46.4
±8.9
53.3
±7.5
76.6 139 (18.3)314 (41.4)9,753 87,565 52.0 90.7 1.6 359 (6.9)504 (9.7)2.6
W. Choi. Near-Online Multi-target Tracking with Aggregated Local Flow Descriptor. In ICCV, 2015.
NOTA
63. using public detections
49.8
±8.3
55.3
±5.4
74.5 136 (17.9)286 (37.7)7,248 83,614 54.1 93.2 1.2 614 (11.3)1,372 (25.3)19.2
L. Chen, H. Ai, R. Chen, Z. Zhuang. Aggregate Tracklet Appearance Features for Multi-Object Tracking. In IEEE Signal Processing Letters, 2019.
oICF
64. online method using public detections
43.2
±10.6
49.3
±9.1
74.3 86 (11.3)368 (48.5)6,651 96,515 47.1 92.8 1.1 381 (8.1)1,404 (29.8)0.4
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.
OTCD_1
65. online method using public detections
44.4
±10.8
45.6
±10.4
75.4 88 (11.6)361 (47.6)5,759 94,927 47.9 93.8 1.0 759 (15.8)1,787 (37.3)17.6
Q. Liu, B. Liu, Y. Wu, W. Li, N. Yu. Real-Time Online Multi-Object Tracking in Compressed Domain. In IEEE Access, 2019.
OVBT
66. online method using public detections
38.4
±8.6
37.8
±5.5
75.4 57 (7.5)359 (47.3)11,517 99,463 45.4 87.8 1.9 1,321 (29.1)2,140 (47.1)0.3
Y. Ban, S. Ba, X. Alameda-Pineda, R. Horaud. Tracking Multiple Persons Based on a Variational Bayesian Model. In BMTT 2016, .
PHD_GSDL16
67. online method using public detections
41.0
±8.9
43.1
±6.9
75.9 86 (11.3)315 (41.5)6,498 99,257 45.6 92.7 1.1 1,810 (39.7)3,650 (80.1)8.3
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.
PMPTracker
68. online method using public detections
40.3
±11.7
38.2
±8.4
73.3 79 (10.4)319 (42.0)10,071 97,524 46.5 89.4 1.7 1,343 (28.9)2,764 (59.4)148.0
Light version of PTZ camera Mutiple People Tracker
PV
69. online method using public detections
50.4
±10.3
50.8
±7.7
77.7 113 (14.9)295 (38.9)2,600 86,780 52.4 97.4 0.4 1,061 (20.2)3,181 (60.7)7.3
X. S. Li, Y. T. Liu, K. F. Wang. Multi-Target Tracking with Trajectory Prediction and Re-Identification//2019 Chinese Automation Congress. IEEE.
QuadMOT16
70. using public detections
44.1
±9.4
38.3
±8.6
76.4 111 (14.6)341 (44.9)6,388 94,775 48.0 93.2 1.1 745 (15.5)1,096 (22.8)1.8
J. Son, M. Baek, M. Cho, B. Han. Multi-Object Tracking with Quadruplet Convolutional Neural Networks. In CVPR, 2017.
TrackerMOTAIDF1MOTPMTMLFPFNRecallPrecisionFAFID Sw.FragHz
RAR16pub
71. online method using public detections
45.9
±9.7
48.8
±7.6
74.8 100 (13.2)318 (41.9)6,871 91,173 50.0 93.0 1.2 648 (13.0)1,992 (39.8)0.9
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.
SCTrack_3L
72. using public detections
36.6
±9.7
36.5
±5.5
73.9 220 (29.0)120 (15.8)35,259 76,653 58.0 75.0 6.0 3,600 (62.1)3,724 (64.3)11.8
Anonymous submission
SDMT
73. online method using public detections
39.6
±8.1
42.3
±6.6
75.5 89 (11.7)373 (49.1)11,130 98,343 46.1 88.3 1.9 602 (13.1)772 (16.8)19.8
M. Thoreau, N. Kottege. Deep Similarity Metric Learning for Real-Time Pedestrian Tracking. In arXiv, 2018.
Seq2Seq
74. using public detections
49.8
±9.5
44.6
±8.0
77.0 114 (15.0)308 (40.6)2,835 87,813 51.8 97.1 0.5 868 (16.7)1,093 (21.1)2.9
Anonymous submission
siameseCos
75. using public detections
49.4
±8.1
49.8
±7.4
75.9 145 (19.1)299 (39.4)6,281 85,384 53.2 93.9 1.1 679 (12.8)823 (15.5)0.8
In preparation
SMOT
76. using public detections
29.7
±8.8
0.0
±0.0
75.2 40 (5.3)362 (47.7)17,426 107,552 41.0 81.1 2.9 3,108 (75.8)4,483 (109.3)0.2
C. Dicle, O. Camps, M. Sznaier. The Way They Move: Tracking Targets with Similar Appearance. In ICCV, 2013.
STAM16
77. online method using public detections
46.0
±9.1
50.0
±8.4
74.9 111 (14.6)331 (43.6)6,895 91,117 50.0 93.0 1.2 473 (9.5)1,422 (28.4)0.2
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.
STRN_MOT16
78. using public detections
48.5
±8.5
53.9
±7.1
73.7 129 (17.0)265 (34.9)9,038 84,178 53.8 91.6 1.5 747 (13.9)2,919 (54.2)13.5
J. Xu, Y. Cao, Z. Zhang, H. Hu. Spatial-Temporal Relation Networks for Multi-Object Tracking. In ICCV, 2019.
S_H_T
79. online method using public detections
49.3
±10.8
50.7
±7.1
77.5 158 (20.8)242 (31.9)16,998 74,142 59.3 86.4 2.9 1,302 (21.9)2,183 (36.8)0.9
Anonymous submission
TARCA
80. online method using public detections
54.8
±12.5
57.3
±8.9
77.4 175 (23.1)291 (38.3)6,679 75,244 58.7 94.1 1.1 487 (8.3)828 (14.1)7.6
Anonymous submission
TrackerMOTAIDF1MOTPMTMLFPFNRecallPrecisionFAFID Sw.FragHz
TBD
81. using public detections
33.7
±8.8
0.0
±0.0
76.5 55 (7.2)411 (54.2)5,804 112,587 38.2 92.3 1.0 2,418 (63.2)2,252 (58.9)1.3
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.
TBSS
82. online method using public detections
44.6
±9.3
42.6
±7.2
75.2 93 (12.3)333 (43.9)4,136 96,128 47.3 95.4 0.7 790 (16.7)1,419 (30.0)3.0
X. Zhou, P. Jiang, Z. Wei, H. Dong, F. Wang. Online Multi-Object Tracking with Structural Invariance Constraint. In BMVC, 2018.
TestUnsup
83. online method using public detections
41.5
±9.0
44.9
±8.8
75.2 104 (13.7)330 (43.5)12,596 93,404 48.8 87.6 2.1 643 (13.2)796 (16.3)19.7
Multi Object Tracking using Deep Structural Cost Minimization in Data Association
TLMHT
84. using public detections
48.7
±8.6
55.3
±5.6
76.4 119 (15.7)338 (44.5)6,632 86,504 52.6 93.5 1.1 413 (7.9)642 (12.2)4.8
H. Sheng, J. Chen, Y. Zhang, W. Ke, Z. Xiong, J. Yu. Iterative Multiple Hypothesis Tracking with Tracklet-level Association. In IEEE Transactions on Circuits and Systems for Video Technology, 2018.
Tracktor++
85. online method using public detections
54.4
±12.0
52.5
±9.6
78.2 144 (19.0)280 (36.9)3,280 79,149 56.6 96.9 0.6 682 (12.1)1,480 (26.2)1.5
P. Bergmann, T. Meinhardt, L. Leal-Taixé. Tracking without bells and whistles. In ICCV, 2019.
Tracktor++v2
86. online method using public detections
56.2
±11.4
54.9
±9.9
79.2 157 (20.7)272 (35.8)2,394 76,844 57.9 97.8 0.4 617 (10.7)1,068 (18.5)1.6
P. Bergmann, T. Meinhardt, L. Leal-Taixé. Tracking without bells and whistles. In ICCV, 2019.
TrajTrack
87. online method using public detections
56.1
±11.2
56.9
±9.0
79.0 163 (21.5)269 (35.4)3,691 75,628 58.5 96.7 0.6 742 (12.7)1,009 (17.2)0.5
Anonymous submission
TrctrD16
88. online method using public detections
54.8
±11.8
53.4
±9.1
77.5 145 (19.1)281 (37.0)2,955 78,765 56.8 97.2 0.5 645 (11.4)1,515 (26.7)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.
UnsupTrack
89. online method using public detections
62.4
±14.6
58.5
±9.8
78.3 205 (27.0)242 (31.9)5,909 61,981 66.0 95.3 1.0 588 (8.9)1,361 (20.6)1.9
Simple Unsupervised Multi-Object Tracking (Under Review)
VAN
90. online method using public detections
57.3
±10.8
57.5
±8.9
79.1 188 (24.8)257 (33.9)3,845 73,489 59.7 96.6 0.6 550 (9.2)772 (12.9)4.7
Anonymous submission
TrackerMOTAIDF1MOTPMTMLFPFNRecallPrecisionFAFID Sw.FragHz
YOONKJ16
91. online method using public detections
47.0
±8.1
50.1
±7.5
75.8 125 (16.5)317 (41.8)7,901 88,179 51.6 92.3 1.3 627 (12.1)945 (18.3)3.5
K. YOON, J. GWAK, Y. SONG, Y. YOON, M. JEON. OneShotDA: Online Multi-object Tracker with One-shot-learning-based Data Association. In IEEE Access, 2020.
SequencesFramesTrajectoriesBoxes
75919759182326

Difficulty Analysis

Sequence difficulty (from easiest to hardest, measured by average MOTA)

MOT16-03

MOT16-03

(53.3 MOTA)

MOT16-06

MOT16-06

(45.0 MOTA)

MOT16-12

MOT16-12

(40.4 MOTA)

...

...

MOT16-08

MOT16-08

(31.4 MOTA)

MOT16-14

MOT16-14

(24.5 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
online method 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.
using public detections This method used the provided detection set as input.
using public detections This method used the provided detection set as input.
new This entry has been submitted or updated less than a week ago.
new This entry has been submitted or updated less than a week ago.

References:


[1] Bernardin, K. & Stiefelhagen, R. Evaluating Multiple Object Tracking Performance: The CLEAR MOT Metrics. Image and Video Processing, 2008(1):1-10, 2008.
[2] Ristani, E., Solera, F., Zou, R., Cucchiara, R. & Tomasi, C. Performance Measures and a Data Set for Multi-Target, Multi-Camera Tracking. In ECCV workshop on Benchmarking Multi-Target Tracking, 2016.
[3] Li, Y., Huang, C. & Nevatia, R. 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.