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

TrackerMOTAIDF1MTMLFPFNID Sw.FragDetector
MHT___ReID
1.
56.4
±0.0
54.239.7% 17.4% 23,79154,1691,478 (21.0)1,547 (22.0)Public
Anonymous submission
Response
2. using public detections
62.0
±0.0
63.837.7% 20.7% 18,30850,039909 (12.5)2,009 (27.7)Public
Anonymous submission
dpt_dpt
3. using public detections
61.3
±0.0
60.432.1% 18.6% 12,41157,481739 (10.8)1,960 (28.6)Public
Anonymous submission
TrkReid
4. using public detections
60.2
±0.0
54.330.7% 20.2% 11,36859,5371,594 (23.7)1,957 (29.1)Public
Anonymous submission
MHT_ReID
5. online method using public detections
27.1
±0.0
36.430.6% 31.4% 13,068118,8291,071 (30.8)1,141 (32.8)Public
Anonymous submission
SCTrack_3L
6. online method
36.6
±0.0
36.529.0% 15.8% 35,25976,6533,600 (62.1)3,724 (64.3)Public
Anonymous submission
DpTrack
7. using public detections
59.3
±0.0
52.827.4% 24.6% 8,56663,6032,045 (31.4)1,555 (23.9)Public
Anonymous submission
Lif_T
8. online method using public detections
61.3
±0.0
64.727.0% 34.0% 4,84465,401389 (6.1)1,034 (16.1)Public
Anonymous submission
ISE_MOT16
9. using public detections
60.1
±0.0
56.926.1% 29.1% 6,96465,044739 (11.5)951 (14.8)Public
MIFT
MPNTrack16
10. online method using public detections
55.9
±0.0
59.926.0% 35.6% 7,08672,902431 (7.2)921 (15.3)Public
Anonymous submission
TrackerMOTAIDF1MTMLFPFNID Sw.FragDetector
TCT5
11. using public detections
51.9
±0.0
51.825.4% 24.9% 15,62270,5571,494 (24.4)2,466 (40.2)Public
Anonymous submission
DS_v2
12.
59.3
±0.0
57.524.2% 29.1% 7,46565,810887 (13.9)2,738 (42.8)Public
Anonymous submission
SORT_Alex
13. online method using public detections
58.4
±0.0
56.324.2% 32.9% 3,43671,846604 (10.0)808 (13.3)Public
Anonymous submission
FFT16
14.
56.5
±0.0
50.123.6% 29.4% 5,83171,8251,635 (27.0)1,607 (26.5)Public
Anonymous submission
ReTrack16
15. online method
57.0
±0.0
54.221.9% 34.3% 4,44673,258688 (11.5)1,543 (25.8)Public
Anonymous submission
HDTR
16. online method using public detections
53.6
±0.0
46.621.2% 37.0% 4,71479,353618 (10.9)833 (14.7)Public
MLT
17. online method
52.8
±0.0
62.621.1% 42.4% 5,36280,444299 (5.4)702 (12.6)Public
Anonymous submission
MCjoint
18. online method using public detections
47.1
±0.0
52.320.4% 46.9% 6,70389,368370 (7.3)598 (11.7)Public
}@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} }
retrack
19. online method using public detections
53.9
±0.0
52.720.3% 32.3% 6,99976,251818 (14.1)2,613 (44.9)Public
Anonymous submission
MOTHP
20.
49.1
±0.0
46.920.0% 38.9% 9,03883,031679 (12.5)850 (15.6)Public
Anonymous submission
TrackerMOTAIDF1MTMLFPFNID Sw.FragDetector
moT_
21. online method using public detections
53.6
±0.0
52.619.9% 32.3% 6,03877,801764 (13.3)2,905 (50.7)Public
Anonymous submission
MOTPP16
22. online method
50.5
±0.0
47.219.6% 39.4% 5,93983,694638 (11.8)823 (15.2)Public
Anonymous submission
CoCT16
23. using public detections
47.5
±0.0
53.219.4% 38.7% 16,97178,044682 (11.9)1,302 (22.8)Public
Anonymous submission
MOTHYPER
24.
50.9
±0.0
47.419.4% 39.4% 4,86684,022619 (11.5)814 (15.1)Public
Anonymous submission
pairwise16
25.
50.0
±0.0
52.419.4% 38.7% 10,99579,568628 (11.1)939 (16.7)Public
Anonymous submission
ENFT16
26.
50.3
±0.0
55.019.2% 39.8% 8,34181,843490 (8.9)754 (13.7)Public
BUAA
HTBT16
27. online method using public detections
50.3
±0.0
55.019.2% 39.8% 8,34181,843490 (8.9)754 (13.7)Public
Anonymous submission
AFN
28. online method using public detections
49.0
±0.0
48.219.1% 35.7% 9,50882,506899 (16.4)1,383 (25.3)Public
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.
FWT
29. using public detections
47.8
±0.0
44.319.1% 38.2% 8,88685,487852 (16.0)1,534 (28.9)Public
R. Henschel, L. Leal-Taixé, D. Cremers, B. Rosenhahn. Fusion of Head and Full-Body Detectors for Multi-Object Tracking. In Trajnet CVPRW, 2018.
MOTHPCLEAN
30.
50.4
±0.0
47.019.1% 39.5% 5,33284,505657 (12.2)862 (16.1)Public
Anonymous submission
TrackerMOTAIDF1MTMLFPFNID Sw.FragDetector
MOTPPF
31. using public detections
48.4
±0.0
48.519.1% 39.8% 9,15284,266595 (11.1)802 (14.9)Public
Anonymous submission
siameseCos
32.
49.4
±0.0
49.819.1% 39.4% 6,28185,384679 (12.8)823 (15.5)Public
In preparation
RTT
33. online method using public detections
49.9
±0.0
49.319.0% 32.8% 9,92780,406955 (17.1)2,247 (40.2)Public
Anonymous submission
Tracktor16
34. using public detections
54.4
±0.0
52.519.0% 36.9% 3,28079,149682 (12.1)1,480 (26.2)Public
P. Bergmann, T. Meinhardt, L. Leal-Taixé. Tracking without bells and whistles. In ICCV, 2019.
MTT_TPR
35. online method using public detections
54.9
±0.0
53.118.7% 34.8% 4,13076,6731,447 (25.0)3,693 (63.7)Public
Anonymous submission
TPM
36. using public detections
51.3
±0.0
47.918.7% 40.8% 2,70185,504569 (10.7)707 (13.3)Public
Anonymous submission
eHAF16
37. online method
47.2
±0.0
52.418.6% 42.8% 12,58683,107542 (10.0)787 (14.5)Public
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.
MOTPP
38. online method using public detections
48.3
±0.0
45.418.6% 40.1% 7,37886,181661 (12.5)834 (15.8)Public
Anonymous submission
CRFTrack16
39. using public detections
50.3
±0.0
54.418.3% 35.7% 7,14882,746702 (12.9)1,387 (25.4)Public
Anonymous submission
CRF_TRACK
40. online method
50.3
±0.0
54.418.3% 35.7% 7,14882,746702 (12.9)1,387 (25.4)Public
Anonymous submission
TrackerMOTAIDF1MTMLFPFNID Sw.FragDetector
NOMT
41. online method using public detections
46.4
±0.0
53.318.3% 41.4% 9,75387,565359 (6.9)504 (9.7)Public
W. Choi. Near-Online Multi-target Tracking with Aggregated Local Flow Descriptor. In ICCV, 2015.
UTA
42. online method using public detections
50.6
±0.0
50.418.3% 33.5% 7,75281,584722 (13.1)2,196 (39.7)Public
Anonymous submission
LMP
43. online method using public detections
48.8
±0.0
51.318.2% 40.1% 6,65486,245481 (9.1)595 (11.3)Public
S. Tang, M. Andriluka, B. Andres, B. Schiele. Multiple People Tracking with Lifted Multicut and Person Re-identification. In CVPR, 2017.
CRF_RNN16
44. online method using public detections
49.0
±0.0
53.918.1% 35.8% 8,49583,838621 (11.5)1,252 (23.2)Public
Anonymous submission
INTERA_MOT
45. online method using public detections
45.4
±0.0
47.718.1% 38.7% 13,40785,547600 (11.3)930 (17.5)Public
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.
MOT_FILTER
46. using public detections
50.2
±0.0
46.817.9% 39.7% 5,26784,812664 (12.4)978 (18.3)Public
Anonymous submission
NOTA
47. online method using public detections
49.8
±0.0
55.317.9% 37.7% 7,24883,614614 (11.3)1,372 (25.3)Public
L. Chen, H. Ai, R. Chen, Z. Zhuang. Aggregate Tracklet Appearance Features for Multi-Object Tracking. In IEEE Signal Processing Letters, 2019.
ENFT
48. online method
50.0
±0.0
54.617.8% 41.1% 8,21482,541479 (8.8)724 (13.2)Public
Anonymous submission
HCC
49. online method using public detections
49.3
±0.0
50.717.8% 39.9% 5,33386,795391 (7.5)535 (10.2)Public
L. Ma, S. Tang, M. Black, L. Gool. Customized Multi-Person Tracker. In Computer Vision -- ACCV 2018, 2018.
DMAN
50. online method using public detections
46.1
±0.0
54.817.4% 42.7% 7,90989,874532 (10.5)1,616 (31.9)Public
J. Zhu, H. Yang, N. Liu, M. Kim, W. Zhang, M. Yang. Online Multi-Object Tracking with Dual Matching Attention Networks. In ECCV, 2018.
TrackerMOTAIDF1MTMLFPFNID Sw.FragDetector
TTL16
51. using public detections
50.4
±0.0
50.117.4% 39.9% 8,49181,156807 (14.5)1,251 (22.5)Public
Anonymous submission
EAGS16
52. using public detections
47.4
±0.0
50.117.3% 42.7% 8,36986,931575 (11.0)913 (17.5)Public
H. Sheng, X. Zhang, Y. Zhang, Y. Wu, J. Chen. Enhanced Association with Supervoxels in Multiple Hypothesis Tracking. In IEEE Access, 2018.
eTC
53. using public detections
49.2
±0.0
56.117.3% 40.3% 8,40083,702606 (11.2)882 (16.3)Public
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.
EDMT
54. using public detections
45.3
±0.0
47.917.0% 39.9% 11,12287,890639 (12.3)946 (18.3)Public
J. Chen, H. Sheng, Y. Zhang, Z. Xiong. Enhancing Detection Model for Multiple Hypothesis Tracking. In BMTT-PETS CVPRw, 2017.
NLLMPa
55. using public detections
47.6
±0.0
47.317.0% 40.4% 5,84489,093629 (12.3)768 (15.0)Public
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.
STRN_MOT16
56. using public detections
48.5
±0.0
53.917.0% 34.9% 9,03884,178747 (13.9)2,919 (54.2)Public
J. Xu, Y. Cao, Z. Zhang, H. Hu. Spatial-Temporal Relation Networks for Multi-Object Tracking. In ICCV, 2019.
CMT16
57. using public detections
49.8
±0.0
59.216.6% 43.6% 9,22981,882365 (6.6)617 (11.2)Public
#Submission: TIP-21190-2019
TLO16
58. using public detections
49.8
±0.0
47.816.6% 40.6% 6,08584,623782 (14.6)1,278 (23.8)Public
Anonymous submission
YOONKJ16
59. online method using public detections
47.0
±0.0
50.116.5% 41.8% 7,90188,179627 (12.1)945 (18.3)Public
Anonymous submission
ASTT
60. using public detections
47.2
±0.0
44.316.3% 41.6% 4,68090,877633 (12.6)814 (16.2)Public
Yi Tao el al., “Adaptive Spatio-temporal Model Based Multiple Object Tracking Considering a Moving Camera[C]”, International Conference on Universal Village (UV), 2018.
TrackerMOTAIDF1MTMLFPFNID Sw.FragDetector
TLO
61. online method using public detections
50.1
±0.0
48.116.3% 40.7% 5,58284,629786 (14.7)1,294 (24.1)Public
Anonymous submission
MHT_DAM
62. online method using public detections
45.8
±0.0
46.116.2% 43.2% 6,41291,758590 (11.9)781 (15.7)Public
C. Kim, F. Li, A. Ciptadi, J. Rehg. Multiple Hypothesis Tracking Revisited. In ICCV, 2015.
MMHT16
63. using public detections
49.9
±0.0
47.316.2% 40.7% 6,11084,455823 (15.3)1,289 (24.0)Public
Anonymous submission
STCG
64. online method
49.3
±0.0
52.016.2% 41.4% 6,88684,979515 (9.6)775 (14.5)Public
Anonymous submission
OMHT16
65. using public detections
49.8
±0.0
46.716.1% 40.4% 6,24484,342888 (16.5)1,332 (24.8)Public
Anonymous submission
KCF16
66. online method using public detections
48.8
±0.0
47.215.8% 38.1% 5,87586,567906 (17.3)1,116 (21.2)Public
P. Chu, H. Fan, C. Tan, H. Ling. Online Multi-Object Tracking with Instance-Aware Tracker and Dynamic Model Refreshment. In WACV, 2019.
TLMHT
67. online method using public detections
48.7
±0.0
55.315.7% 44.5% 6,63286,504413 (7.9)642 (12.2)Public
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.
deepS2
68. using public detections
46.0
±0.0
46.515.5% 42.6% 5,12492,697693 (14.1)759 (15.4)Public
ID 32
JMC
69. online method
46.3
±0.0
46.315.5% 39.7% 6,37390,914657 (13.1)1,114 (22.2)Public
S. Tang, B. Andres, M. Andriluka, B. Schiele. Multi-Person Tracking by Multicuts and Deep Matching. In BMTT, 2016.
SCNet
70. using public detections
50.0
±0.0
51.115.5% 34.1% 10,52679,755866 (15.4)2,141 (38.1)Public
Anonymous submission
TrackerMOTAIDF1MTMLFPFNID Sw.FragDetector
SRPN
71. using public detections
44.0
±0.0
36.615.5% 45.7% 18,78482,3181,047 (19.1)1,118 (20.4)Public
Anonymous submission
AEb
72. using public detections
42.9
±0.0
48.715.3% 49.0% 4,48799,310375 (8.2)1,334 (29.3)Public
Anonymous submission
AOReid
73. using public detections
48.2
±0.0
50.815.3% 36.8% 10,28383,301821 (15.1)1,963 (36.1)Public
Anonymous submission
RFS
74. online method using public detections
49.0
±0.0
50.315.3% 43.0% 9,95382,394701 (12.8)1,663 (30.3)Public
Anonymous submission
DAST
75. online method using public detections
48.9
±0.0
53.215.2% 36.2% 9,98782,427838 (15.3)1,936 (35.3)Public
Anonymous submission
MOTDT
76. using public detections
47.6
±0.0
50.915.2% 38.3% 9,25385,431792 (14.9)1,858 (35.0)Public
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.
DeepMP16
77. using public detections
48.7
±0.0
50.115.0% 43.6% 4,11188,862535 (10.4)873 (17.0)Public
Anonymous submission
MEN
78. online method using public detections
50.0
±0.0
52.815.0% 37.0% 6,11784,271706 (13.1)1,797 (33.4)Public
Anonymous submission
MHT_bLSTM6
79. using public detections
42.1
±0.0
47.814.9% 44.4% 11,63793,172753 (15.4)1,156 (23.6)Public
C. Kim, F. Li, J. Rehg. Multi-object Tracking with Neural Gating Using Bilinear LSTM. In ECCV, 2018.
PV
80. online method
50.4
±0.0
50.814.9% 38.9% 2,60086,7801,061 (20.2)3,181 (60.7)Public
Anonymous submission
TrackerMOTAIDF1MTMLFPFNID Sw.FragDetector
WOLF
81. online method using public detections
47.5
±0.0
41.914.8% 35.7% 10,93983,4281,276 (23.5)2,480 (45.7)Public
Anonymous submission
QuadMOT16
82.
44.1
±0.0
38.314.6% 44.9% 6,38894,775745 (15.5)1,096 (22.8)Public
J. Son, M. Baek, M. Cho, B. Han. Multi-Object Tracking with Quadruplet Convolutional Neural Networks. In CVPR, 2017.
STAM16
83. online method
46.0
±0.0
50.014.6% 43.6% 6,89591,117473 (9.5)1,422 (28.4)Public
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.
JCSTD
84. using public detections
47.4
±0.0
41.114.4% 36.4% 8,07686,6381,266 (24.1)2,697 (51.4)Public
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.
SRPN16
85. online method using public detections
48.2
±0.0
51.314.2% 36.8% 7,76785,973790 (14.9)2,006 (38.0)Public
Anonymous submission
DCCRF16
86. online method
44.8
±0.0
39.714.1% 42.3% 5,61394,133968 (20.0)1,378 (28.5)Public
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
87. online method using public detections
46.2
±0.0
49.414.1% 44.0% 5,12692,367598 (12.1)1,127 (22.8)Public
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.
MTDF
88. using public detections
45.7
±0.0
40.114.1% 36.4% 12,01884,9701,987 (37.2)3,377 (63.2)Public
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.
AMIR
89. online method using public detections
47.2
±0.0
46.314.0% 41.6% 2,68192,856774 (15.8)1,675 (34.1)Public
A. Sadeghian, A. Alahi, S. Savarese. Tracking The Untrackable: Learning To Track Multiple Cues with Long-Term Dependencies. In ICCV, 2017.
TestUnsup
90. using public detections
41.5
±0.0
44.913.7% 43.5% 12,59693,404643 (13.2)796 (16.3)Public
Multi Object Tracking using Deep Structural Cost Minimization in Data Association
TrackerMOTAIDF1MTMLFPFNID Sw.FragDetector
LSST16O
91. online method using public detections
49.2
±0.0
56.513.4% 41.4% 7,18784,875606 (11.3)2,497 (46.7)Public
Anonymous submission
LFNF16
92. using public detections
43.6
±0.0
41.613.3% 45.7% 6,61695,363836 (17.5)938 (19.7)Public
Sheng H, Hao L, Chen J, et al. Robust Local Effective Matching Model for Multi-Target Tracking. In PCM, 2017
RAR16pub
93. using public detections
45.9
±0.0
48.813.2% 41.9% 6,87191,173648 (13.0)1,992 (39.8)Public
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.
GCRA
94. online method using public detections
48.2
±0.0
48.612.9% 41.1% 5,10488,586821 (16.0)1,117 (21.7)Public
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.
TBSS
95. online method using public detections
44.6
±0.0
42.612.3% 43.9% 4,13696,128790 (16.7)1,419 (30.0)Public
X. Zhou, P. Jiang, Z. Wei, H. Dong, F. Wang. Online Multi-Object Tracking with Structural Invariance Constraint. In BMVC, 2018.
SDMT
96. online method using public detections
39.6
±0.0
42.311.7% 49.1% 11,13098,343602 (13.1)772 (16.8)Public
M. Thoreau, N. Kottege. Deep Similarity Metric Learning for Real-Time Pedestrian Tracking. In arXiv, 2018.
LINF1
97. using public detections
41.0
±0.0
45.711.6% 51.3% 7,89699,224430 (9.4)963 (21.1)Public
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.
OTCD_1
98. using public detections
44.4
±0.0
45.611.6% 47.6% 5,75994,927759 (15.8)1,787 (37.3)Public
Q. Liu, B. Liu, Y. Wu, W. Li, N. Yu. Real-Time Online Multi-Object Tracking in Compressed Domain. In IEEE Access, 2019.
oICF
99. online method using public detections
43.2
±0.0
49.311.3% 48.5% 6,65196,515381 (8.1)1,404 (29.8)Public
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.
PHD_GSDL16
100. using public detections
41.0
±0.0
43.111.3% 41.5% 6,49899,2571,810 (39.7)3,650 (80.1)Public
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.
TrackerMOTAIDF1MTMLFPFNID Sw.FragDetector
GMPHD_ReId
101.
40.4
±0.0
49.711.2% 43.3% 6,572101,266792 (17.8)2,529 (56.9)Public
N. Baisa. Occlusion-robust Online Multi-object Visual Tracking using a GM-PHD Filter with a CNN-based Re-identification. In , 2019.
PHD_T
102. online method
40.4
±0.0
49.711.2% 43.3% 6,572101,266792 (17.8)2,529 (56.9)Public
Anonymous submission
CDA_DDALv2
103. online method using public detections
43.9
±0.0
45.110.7% 44.4% 6,45095,175676 (14.1)1,795 (37.6)Public
S. Bae and K. Yoon, Confidence-Based Data Association and Discriminative Deep Appearance Learning for Robust Online Multi-Object Tracking , In IEEE TPAMI, 2017.
OST16
104.
41.5
±0.0
39.110.7% 45.6% 5,91999,7091,056 (23.3)1,487 (32.8)Public
Anonymous submission
PMPTracker
105. online method
40.3
±0.0
38.210.4% 42.0% 10,07197,5241,343 (28.9)2,764 (59.4)Public
Light version of PTZ camera Mutiple People Tracker
FTPLS
106. using public detections
38.2
±0.0
47.59.6% 44.0% 18,91593,051689 (14.1)2,006 (41.0)Public
Anonymous submission
LTTSC-CRF
107. online method using public detections
37.6
±0.0
42.19.6% 55.2% 11,969101,343481 (10.8)1,012 (22.8)Public
N. Le, A. Heili, M. Odobez. Long-Term Time-Sensitive Costs for CRF-Based Tracking by Detection. In ECCVw, 2016.
D_cost16
108. online method using public detections
39.9
±0.0
35.38.7% 50.2% 1,133107,586790 (19.3)824 (20.1)Public
Anonymous submission
LP2D
109. using public detections
35.7
±0.0
34.28.7% 50.7% 5,084111,163915 (23.4)1,264 (32.4)Public
MOT baseline: Linear programming on 2D image coordinates.
GMMCP
110. online method using public detections
38.1
±0.0
35.58.6% 50.9% 6,607105,315937 (22.2)1,669 (39.5)Public
A. Dehghan, S. Assari, M. Shah.. GMMCP-Tracker:Globally Optimal Generalized Maximum Multi Clique Problem for Multiple Object Tracking. In CVPR, 2015.
TrackerMOTAIDF1MTMLFPFNID Sw.FragDetector
GoturnM16
111. online method
37.5
±0.0
25.18.4% 46.5% 17,74692,8673,277 (66.8)2,994 (61.0)Public
Anonymous submission
EAMTT_pub
112. online method using public detections
38.8
±0.0
42.47.9% 49.1% 8,114102,452965 (22.0)1,657 (37.8)Public
R. Sanchez-Matilla, F. Poiesi, A. Cavallaro "Multi-target tracking with strong and weak detections" in BMTT ECCVw 2016
RNN_A_P
113. online method using public detections
34.0
±0.0
33.77.9% 51.0% 8,562109,2692,479 (61.9)3,393 (84.7)Public
Anonymous submission
CEM
114. using public detections
33.2
±0.0
0.07.8% 54.4% 6,837114,322642 (17.2)731 (19.6)Public
A. Milan, S. Roth, K. Schindler. Continuous Energy Minimization for Multitarget Tracking. In IEEE TPAMI, 2014.
HAM_ACT16
115. using public detections
38.1
±0.0
43.37.8% 54.4% 6,976105,434418 (9.9)707 (16.8)Public
HISP_T
116. using public detections
35.9
±0.0
28.97.8% 50.1% 6,412107,9182,594 (63.6)2,298 (56.3)Public
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.
HISP_DAL
117. using public detections
37.4
±0.0
30.57.6% 50.9% 3,222108,8652,101 (52.1)2,151 (53.4)Public
N. Baisa. Robust Online Multi-target Visual Tracking using a HISP Filter with Discriminative Deep Appearance Learning. In CoRR, 2019.
JCmin_MOT
118. online method using public detections
36.7
±0.0
36.27.5% 54.4% 2,936111,890667 (17.3)831 (21.5)Public
M. Abhijeet Boragule. Joint Cost Minimization for Multi-Object Tracking. In 2017 IEEE International Conference on Advanced Vide and Signale Based Surveillance, 2017.
OVBT
119. online method using public detections
38.4
±0.0
37.87.5% 47.3% 11,51799,4631,321 (29.1)2,140 (47.1)Public
Y. Ban, S. Ba, X. Alameda-Pineda, R. Horaud. Tracking Multiple Persons Based on a Variational Bayesian Model. In BMTT 2016, .
TBD
120. using public detections
33.7
±0.0
0.07.2% 54.2% 5,804112,5872,418 (63.2)2,252 (58.9)Public
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.
TrackerMOTAIDF1MTMLFPFNID Sw.FragDetector
AM_ADM
121. using public detections
40.1
±0.0
43.87.1% 46.2% 8,50399,891789 (17.5)1,736 (38.4)Public
S. Lee, M. Kim, S. Bae, Learning Discriminative Appearance Models for Online Multi-Object Tracking with Appearance Discriminability Measures, In IEEE Access, 2018.
GM_PHD_Dl
122. using public detections
34.3
±0.0
20.57.1% 51.5% 2,350111,8865,605 (145.1)5,357 (138.7)Public
Anonymous submission
GM_PHD_DAL
123. using public detections
35.1
±0.0
26.67.0% 51.4% 2,350111,8864,047 (104.8)5,338 (138.2)Public
N. Baisa. Online Multi-object Visual Tracking using a GM-PHD Filter with Deep Appearance Learning. In 22nd International Conference on Information Fusion, 2019.
GM_PHD_e17
124. using public detections
33.8
±0.0
25.36.3% 54.9% 1,766115,1303,778 (102.5)3,874 (105.1)Public
Anonymous submission
KVIOU16
125. online method
33.4
±0.0
32.65.9% 59.6% 2,764117,971760 (21.5)1,473 (41.7)Public
Anonymous submission
GM_PHD_N1T
126. online method using public detections
33.3
±0.0
25.55.5% 56.0% 1,750116,4523,499 (96.8)3,594 (99.5)Public
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.
SMOT
127. online method using public detections
29.7
±0.0
0.05.3% 47.7% 17,426107,5523,108 (75.8)4,483 (109.3)Public
C. Dicle, O. Camps, M. Sznaier. The Way They Move: Tracking Targets with Similar Appearance. In ICCV, 2013.
GMPHD_HDA
128. using public detections
30.5
±0.0
33.44.6% 59.7% 5,169120,970539 (16.0)731 (21.7)Public
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.
CppSORT
129. online method
31.5
±0.0
27.74.3% 59.9% 3,048120,2781,587 (46.6)2,239 (65.8)Public
S. Murray. Real-Time Multiple Object Tracking - A Study on the Importance of Speed. In arXiv preprint arXiv:1709.03572, 2017.
DP_NMS
130. online method using public detections
26.2
±0.0
31.24.1% 67.5% 3,689130,557365 (12.9)638 (22.5)Public
H. Pirsiavash, D. Ramanan, C. Fowlkes. Globally-Optimal Greedy Algorithms for Tracking a Variable Number of Objects. In CVPR, 2011.
TrackerMOTAIDF1MTMLFPFNID Sw.FragDetector
JPDA_m
131. online method
26.2
±0.0
0.04.1% 67.5% 3,689130,549365 (12.9)638 (22.5)Public
H. Rezatofighi, A. Milan, Z. Zhang, Q. Shi, A. Dick, I. Reid. Joint Probabilistic Data Association Revisited. In ICCV, 2015.
DCOR
132. online method using public detections
28.3
±0.0
21.73.4% 63.9% 1,618128,345849 (28.7)2,592 (87.5)Public
Anonymous submission
test_trker
133. using public detections
0.0
±0.0
0.00.0% 100.0% 7182,3260 (nan)0 (nan)Public
Anonymous submission
SequencesFramesTrajectoriesBoxes
75919759182326

Difficulty Analysis

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

MOT16-01

MOT16-01

(0.0% MOTA)

MOT16-03

MOT16-03

(0.0% MOTA)

MOT16-06

MOT16-06

(0.0% MOTA)

...

...

MOT16-12

MOT16-12

(0.0% MOTA)

MOT16-14

MOT16-14

(0.0% MOTA)


Evaluation Measures

Lower is better. Higher is better.
Measure Better Perfect Description
Avg Rank lower 1 This is the rank of each tracker averaged over all present evaluation measures.
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.

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.
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.