MOT17 Results

Download all results for this benchmark

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


Showing only entries that use public detections!

TrackerAvg RankMOTA IDF1MTMLFPFNID Sw.FragHzDetector
MOT_BJ
1. online method using public detections 		81.0
-7.3
±23.5
1.40.0% 99.1% 52,007548,5314,824 (1,734.0)8,621 (3,098.8)0.0Public
Anonymous submission
baitrack
2. using public detections 		48.3
37.6
±19.4
20.321.0% 30.9% 99,085244,0018,808 (155.2)6,708 (118.2)6.4Public
Anonymous submission
XYHv2
3. online method using public detections 		75.5
39.9
±12.4
23.89.9% 41.8% 29,713296,70412,900 (272.1)12,911 (272.3)7.8Public
Anonymous submission
GLMBS3
4. using public detections 		68.5
38.0
±13.7
32.39.3% 52.8% 38,874304,0166,963 (151.0)3,927 (85.2)4.9Public
Anonymous submission
ZM
5. online method using public detections 		65.9
43.5
±13.9
32.614.5% 39.9% 25,083284,4059,197 (185.4)8,849 (178.4)14.4Public
Anonymous submission
TM_track
6. online method using public detections 		75.3
41.1
±14.9
32.813.2% 41.3% 27,606287,51117,408 (355.0)15,197 (309.9)2.5Public
Anonymous submission
GM_PHD
7. online method using public detections 		62.4
42.1
±13.0
33.911.9% 42.7% 18,214297,64610,698 (226.4)10,864 (229.9)9.9Public
Anonymous submission
GMPHD_N1Tr
8. online method using public detections 		63.1
42.1
±13.5
33.911.9% 42.7% 18,214297,64610,698 (226.4)10,864 (229.9)9.9Public
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.
GM_PHD
9. online method using public detections 		65.3
36.4
±14.1
33.94.1% 57.3% 23,723330,7674,607 (111.3)11,317 (273.5)38.4Public
V. Eiselein, D. Arp, M. Pätzold, T. Sikora. Real-time Multi-Human Tracking using a Probability Hypothesis Density Filter and multiple detectors. In 9th IEEE International Conference on Advanced Video and Signal-Based Surveillance, 2012.
dcor
10. online method using public detections 		53.2
45.0
±14.2
34.015.4% 38.2% 30,231275,2654,801 (93.7)8,498 (165.9)44.4Public
Anonymous submission
TrackerAvg RankMOTA IDF1MTMLFPFNID Sw.FragHzDetector
GM_PHD_D
11. online method using public detections 		59.2
44.0
±13.8
34.214.8% 39.4% 19,135283,53013,556 (272.5)13,821 (277.8)9.9Public
Anonymous submission
DCORV2
12. online method using public detections 		49.9
45.5
±13.9
36.114.6% 40.4% 21,161282,9013,592 (72.0)7,696 (154.4)35.5Public
Anonymous submission
GMPHD_DAL
13. online method using public detections 		60.8
44.4
±13.9
36.214.9% 39.4% 19,170283,38011,137 (223.7)13,900 (279.3)3.4Public
N. Baisa. Online Multi-object Visual Tracking using a GM-PHD Filter with Deep Appearance Learning. In 22nd International Conference on Information Fusion, 2019.
GMPHD_KCF
14. online method using public detections 		70.8
39.6
±13.6
36.68.8% 43.3% 50,903284,2285,811 (117.1)7,414 (149.4)3.3Public
T. Kutschbach, E. Bochinski, V. Eiselein, T. Sikora. Sequential Sensor Fusion Combining Probability Hypothesis Density and Kernelized Correlation Filters for Multi-Object Tracking in Video Data. In International Workshop on Traffic and Street Surveillance for Safety and Security at IEEE AVSS 2017, 2017.
98K
15. using public detections new 		52.2
40.8
±17.2
37.015.6% 38.1% 32,312298,1743,514 (74.5)4,991 (105.8)17.7Public
Anonymous submission
MOTbyReID
16. online method using public detections 		63.0
43.6
±13.7
37.117.6% 36.3% 35,725270,03612,347 (236.8)11,408 (218.8)2.5Public
Anonymous submission
GOTURN_3B
17. online method using public detections 		57.7
44.3
±13.7
38.513.0% 43.2% 30,302279,1444,861 (96.2)5,277 (104.4)48.6Public
Anonymous submission
HISP_T17
18. online method using public detections 		59.6
44.6
±14.2
38.815.1% 38.8% 25,478276,39510,617 (208.1)7,487 (146.8)4.7Public
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.
GF
19. online method using public detections 		54.5
45.0
±13.9
39.115.0% 39.0% 22,387277,33510,397 (204.5)7,421 (145.9)9.9Public
Anonymous submission
GMPHD_SHA
20. online method using public detections 		57.8
43.7
±12.5
39.211.7% 43.0% 25,935287,7583,838 (78.3)5,056 (103.2)9.2Public
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.
TrackerAvg RankMOTA IDF1MTMLFPFNID Sw.FragHzDetector
IOU17
21. using public detections 		51.5
45.5
±13.6
39.415.7% 40.5% 19,993281,6435,988 (119.6)7,404 (147.8)1,522.9Public
E. Bochinski, V. Eiselein, T. Sikora. High-Speed Tracking-by-Detection Without Using Image Information. In International Workshop on Traffic and Street Surveillance for Safety and Security at IEEE AVSS 2017, 2017.
SORT17
22. online method using public detections 		63.0
43.1
±13.3
39.812.5% 42.3% 28,398287,5824,852 (99.0)7,127 (145.4)143.3Public
A. Bewley, Z. Ge, L. Ott, F. Ramos, B. Upcroft. Simple online and realtime tracking. In 2016 IEEE International Conference on Image Processing (ICIP), 2016.
HISP_DAL17
23. online method using public detections 		57.7
45.4
±13.9
39.914.8% 39.2% 21,820277,4738,727 (171.7)7,147 (140.6)3.2Public
N. Baisa. Robust Online Multi-target Visual Tracking using a HISP Filter with Discriminative Deep Appearance Learning. In CoRR, 2019.
zxbtk17
24. online method using public detections 		53.6
45.1
±14.7
40.017.7% 31.8% 33,186273,5313,303 (64.1)8,148 (158.1)8.3Public
Anonymous submission
ms_dh
25. online method using public detections 		69.5
42.6
±14.6
40.113.6% 40.0% 31,878284,5287,446 (150.2)14,736 (297.3)4.0Public
Anonymous submission
c3d_Track
26. online method using public detections 		62.8
41.5
±13.7
40.210.7% 48.5% 33,332292,9313,890 (80.9)11,454 (238.2)22.2Public
Anonymous submission
EDA_GNN
27. online method using public detections 		49.3
45.5
±13.8
40.515.6% 40.6% 25,685277,6634,091 (80.5)5,579 (109.8)39.3Public
Paper ID 2713
QiMOT
28. online method using public detections 		55.5
47.2
±13.1
40.815.5% 39.9% 18,907274,8284,320 (84.2)5,917 (115.4)1.8Public
Anonymous submission
TAR_1
29. online method using public detections 		39.9
51.6
±11.9
41.421.7% 28.7% 33,514235,8593,629 (62.4)5,949 (102.2)5.6Public
Anonymous submission
YoloSort
30. online method using public detections 		55.7
29.5
±24.1
41.715.0% 36.4% 154,747238,2414,888 (84.6)4,952 (85.7)14.4Public
Anonymous submission
TrackerAvg RankMOTA IDF1MTMLFPFNID Sw.FragHzDetector
EAMTT
31. online method using public detections 		58.9
42.6
±13.3
41.812.7% 42.7% 30,711288,4744,488 (91.8)5,720 (117.0)12.0Public
R. Sanchez-Matilla, F. Poiesi, A. Cavallaro. Online Multi-target Tracking with Strong and Weak Detections. In Computer Vision -- ECCV 2016 Workshops, 2016.
OTCD_1
32. online method using public detections 		52.7
44.9
±13.6
42.314.0% 44.2% 16,280291,1363,573 (73.8)5,444 (112.5)46.5Public
Q. Liu, B. Liu, Y. Wu, W. Li, N. Yu. Real-Time Online Multi-Object Tracking in Compressed Domain. In IEEE Access, 2019.
PHD_LMP
33. online method using public detections 		54.6
45.9
±13.1
42.515.5% 37.9% 27,946272,1964,977 (96.2)6,985 (135.0)29.4Public
Anonymous submission
DeepMOTRPN
34. online method using public detections 		44.3
48.1
±14.5
43.017.6% 38.6% 26,490262,5783,696 (69.1)5,353 (100.1)4.9Public
Anonymous submission
PHD_GM
35. online method using public detections 		43.4
48.8
±13.4
43.219.1% 35.2% 26,260257,9714,407 (81.2)6,448 (118.8)22.3Public
R. Sanchez-Matilla, A. Cavallaro. A predictor of moving objects for First-Person vision. In Proceedings of IEEE International Conference Image Processing, 2019.
DAN__test
36. using public detections 		65.0
43.0
±14.7
43.313.5% 40.0% 30,367283,5337,576 (152.3)14,990 (301.3)1.8Public
Anonymous submission
SNM17
37. online method using public detections 		58.7
46.8
±13.8
43.416.2% 37.1% 25,104271,0424,213 (81.1)9,891 (190.3)0.8Public
Anonymous submission
Q_ls
38. online method using public detections 		46.8
50.2
±14.4
43.619.7% 37.3% 23,143253,1514,414 (80.1)6,112 (110.9)1.8Public
Anonymous submission
DSA_MOT17
39. online method using public detections 		42.5
45.0
±12.6
43.615.8% 39.2% 21,442286,4822,491 (50.6)3,824 (77.7)9.9Public
Anonymous submission
DualAtte
40. online method using public detections 		46.3
48.4
±14.5
43.717.6% 39.0% 24,915262,6543,423 (64.0)5,192 (97.1)0.3Public
Anonymous submission
TrackerAvg RankMOTA IDF1MTMLFPFNID Sw.FragHzDetector
CGHA_MOT
41. online method using public detections 		63.2
41.2
±14.1
44.08.3% 45.9% 25,462299,1127,294 (155.2)18,655 (397.0)11.4Public
Anonymous submission
AEb_O
42. online method using public detections 		43.9
46.4
±13.9
44.916.5% 41.4% 17,030283,2652,266 (45.5)5,053 (101.5)1.8Public
Anonymous submission
LT17
43. online method using public detections 		49.1
47.7
±16.6
45.217.3% 36.0% 27,856263,0624,042 (75.7)9,183 (172.0)7.2Public
Anonymous submission
MTDF17
44. online method using public detections 		50.8
49.6
±13.9
45.218.9% 33.1% 37,124241,7685,567 (97.4)9,260 (162.0)1.2Public
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.
SOTD_MC
45. online method using public detections 		39.7
48.4
±15.0
45.519.4% 35.9% 33,525255,0912,531 (46.2)4,944 (90.2)67.0Public
Anonymous submission
MASS
46. online method using public detections 		49.5
46.9
±14.1
46.016.9% 36.3% 25,733269,1164,478 (85.6)11,994 (229.3)17.1Public
Anonymous submission
D_SST_V1
47. online method using public detections 		62.3
42.7
±13.9
46.111.8% 44.4% 18,861298,9895,531 (117.7)13,775 (293.0)2.3Public
Anonymous submission
CASC_MOT
48. online method using public detections 		57.1
42.3
±12.8
46.89.1% 44.1% 21,035300,7973,616 (77.4)16,656 (356.7)11.4Public
Anonymous submission
MHT_ReID7
49. using public detections 		52.6
46.5
±13.7
46.918.8% 40.3% 22,203276,3743,386 (66.4)8,521 (167.0)1.6Public
Anonymous submission
AFN17
50. using public detections 		28.8
51.5
±13.0
46.920.6% 35.5% 22,391248,4202,593 (46.3)4,308 (77.0)1.8Public
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.
TrackerAvg RankMOTA IDF1MTMLFPFNID Sw.FragHzDetector
AEb
51. using public detections 		31.4
47.9
±13.6
47.018.1% 40.7% 15,828276,1792,082 (40.8)4,733 (92.7)66.9Public
Anonymous submission
GMPHDOGM17
52. online method using public detections 		32.8
49.9
±13.6
47.119.7% 38.0% 24,024255,2773,125 (57.1)3,540 (64.6)30.7Public
Y. Song, K. Yoon, Y. Yoon, K. Yow, M. Jeon. Online Multi-Object Tracking Framework with the GMPHD Filter and Occlusion Group Management. In arXiv:1907.13347, 2019.
MHT_DAM
53. using public detections 		35.8
50.7
±13.7
47.220.8% 36.9% 22,875252,8892,314 (41.9)2,865 (51.9)0.9Public
C. Kim, F. Li, A. Ciptadi, J. Rehg. Multiple Hypothesis Tracking Revisited. In ICCV, 2015.
FWT
54. using public detections 		32.9
51.3
±13.1
47.621.4% 35.2% 24,101247,9212,648 (47.2)4,279 (76.3)0.2Public
R. Henschel, L. Leal-Taixé, D. Cremers, B. Rosenhahn. Fusion of Head and Full-Body Detectors for Multi-Object Tracking. In Trajnet CVPRW, 2018.
Lab031
55. using public detections 		48.4
46.9
±16.2
48.117.7% 36.1% 31,634263,9383,795 (71.3)10,498 (197.3)9.4Public
Anonymous submission
TPbase17
56. online method using public detections 		57.4
43.3
±15.0
48.216.2% 36.6% 49,992265,8154,194 (79.3)12,103 (228.8)22.2Public
Anonymous submission
FPSN
57. online method using public detections 		52.8
44.9
±13.9
48.416.5% 35.8% 33,757269,9527,136 (136.8)14,491 (277.8)10.1Public
S. Lee, E. Kim. Multiple Object Tracking via Feature Pyramid Siamese Networks. In IEEE ACCESS, 2018.
HDTR
58. using public detections 		21.3
54.1
±11.4
48.423.3% 34.8% 18,002238,8181,895 (32.9)2,693 (46.7)1.8Public
SiaIOU
59. using public detections 		43.3
48.5
±16.7
48.518.9% 38.8% 26,867260,2783,152 (58.5)4,391 (81.5)8.3Public
Anonymous submission
FAMNet
60. online method using public detections 		32.8
52.0
±12.0
48.719.1% 33.4% 14,138253,6163,072 (55.8)5,318 (96.6)0.0Public
P. Chu, H. Ling. FAMNet: Joint Learning of Feature, Affinity and Multi-dimensional Assignment for Online Multiple Object Tracking. In ICCV, 2019.
TrackerAvg RankMOTA IDF1MTMLFPFNID Sw.FragHzDetector
MOT_HY
61. using public detections 		50.0
47.3
±121.2
49.417.2% 33.8% 46,875246,0614,231 (75.0)8,188 (145.2)2.0Public
Anonymous submission
OLGT_new
62. online method using public detections 		47.6
45.7
±22.8
49.410.8% 75.5% 6,915298,2881,418 (30.1)3,641 (77.2)6.1Public
Anonymous submission
PHD_GSDL17
63. online method using public detections 		46.8
48.0
±13.6
49.617.1% 35.6% 23,199265,9543,998 (75.6)8,886 (168.1)6.7Public
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.
MOT_TBC
64. using public detections 		30.0
53.9
±15.7
50.020.2% 36.7% 24,584232,6702,945 (50.1)4,612 (78.5)6.7Public
Anonymous submission
MFT
65. online method using public detections 		43.7
53.1
±16.6
50.120.4% 39.4% 35,295225,6063,681 (61.3)6,271 (104.5)0.7Public
Anonymous submission
OST
66. using public detections 		42.4
49.7
±14.0
50.417.0% 36.7% 21,811258,6493,077 (56.8)4,339 (80.1)1.7Public
Anonymous submission
JBNOT
67. using public detections 		31.2
52.6
±12.3
50.819.7% 35.8% 31,572232,6593,050 (51.9)3,792 (64.5)5.4Public
R. Henschel, Y. Zou, B. Rosenhahn. Multiple People Tracking using Body and Joint Detections. In CVPRW, 2019.
DS_TW_F
68. online method using public detections 		38.3
45.7
±27.0
50.910.8% 75.4% 6,528298,3681,329 (28.2)3,180 (67.5)66.9Public
Anonymous submission
HAM_SADF17
69. online method using public detections 		37.6
48.3
±13.2
51.117.1% 41.7% 20,967269,0381,871 (35.8)3,020 (57.7)5.0Public
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.
EDMT17
70. using public detections 		34.7
50.0
±13.9
51.321.6% 36.3% 32,279247,2972,264 (40.3)3,260 (58.0)0.6Public
J. Chen, H. Sheng, Y. Zhang, Z. Xiong. Enhancing Detection Model for Multiple Hypothesis Tracking. In BMTT-PETS CVPRw, 2017.
TrackerAvg RankMOTA IDF1MTMLFPFNID Sw.FragHzDetector
DGCT
71. using public detections 		20.5
54.5
±13.1
51.321.0% 35.4% 10,471243,1432,865 (50.3)4,889 (85.9)7.0Public
CJY, HYW, KHW @ HRI-SH
PV
72. online method using public detections 		34.8
52.8
±14.1
51.819.7% 34.0% 15,884246,9393,711 (66.0)8,757 (155.7)3.5Public
Anonymous submission
MHT_bLSTM
73. using public detections 		42.7
47.5
±12.6
51.918.2% 41.7% 25,981268,0422,069 (39.4)3,124 (59.5)1.9Public
C. Kim, F. Li, J. Rehg. Multi-object Tracking with Neural Gating Using Bilinear LSTM. In ECCV, 2018.
AM_ADM17
74. online method using public detections 		39.4
48.1
±13.8
52.113.4% 39.7% 25,061265,4952,214 (41.8)5,027 (94.9)5.7Public
S. Lee, M. Kim, S. Bae, Learning Discriminative Appearance Models for Online Multi-Object Tracking with Appearance Discriminability Measures, In IEEE Access, 2018.
UTA
75. online method using public detections 		30.5
53.0
±11.6
52.221.7% 31.5% 24,468238,3562,292 (39.7)6,231 (107.9)5.0Public
Anonymous submission
DeepMP17
76. using public detections 		29.1
50.4
±13.1
52.318.8% 38.7% 22,535255,3561,868 (34.1)3,473 (63.4)7.4Public
Tracktor17
77. online method using public detections 		31.9
53.5
±14.5
52.319.5% 36.6% 12,201248,0472,072 (37.0)4,611 (82.3)1.5Public
P. Bergmann, T. Meinhardt, L. Leal-Taixé. Tracking without bells and whistles. In ICCV, 2019.
TPM
78. using public detections 		27.1
54.2
±13.0
52.622.8% 37.5% 13,739242,7301,824 (32.0)2,472 (43.4)0.8Public
Anonymous submission
TppNoFPN
79. using public detections 		35.6
52.4
±15.3
52.618.5% 37.2% 18,635247,1042,726 (48.5)5,461 (97.2)4.2Public
Anonymous submission
MOTDT17
80. online method using public detections 		32.3
50.9
±11.9
52.717.5% 35.7% 24,069250,7682,474 (44.5)5,317 (95.7)18.3Public
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.
TrackerAvg RankMOTA IDF1MTMLFPFNID Sw.FragHzDetector
2MPT
81. using public detections 		35.7
48.1
±14.2
52.917.4% 39.6% 30,650260,1331,860 (34.5)2,784 (51.7)2.7Public
Anonymous submission
TOPA
82. online method using public detections 		30.2
51.8
±13.5
53.419.6% 33.1% 27,603241,5462,668 (46.7)5,790 (101.2)443.9Public
Anonymous submission
DTBasline
83. online method using public detections 		28.6
51.1
±11.7
53.416.7% 35.5% 20,309253,2452,549 (46.2)5,910 (107.2)22.2Public
Anonymous submission
MOT17ZH
84. online method using public detections 		36.4
51.1
±13.7
53.416.7% 35.5% 20,309253,2452,549 (46.2)5,910 (107.2)3.7Public
Anonymous submission
SNet_pub
85. online method using public detections new 		37.9
51.7
±12.0
53.418.0% 33.5% 26,809243,0662,735 (48.0)6,157 (108.2)2.5Public
Anonymous submission
SRPN17
86. online method using public detections 		37.5
51.0
±11.7
53.516.8% 35.1% 21,011252,8082,596 (47.0)5,981 (108.4)4.1Public
Anonymous submission
DEEP_TAMA
87. online method using public detections 		32.8
50.3
±13.3
53.519.2% 37.5% 25,479252,9962,192 (39.7)3,978 (72.1)1.5Public
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.
LSMT
88. online method using public detections 		27.7
51.9
±12.0
53.517.4% 35.0% 18,672250,6622,257 (40.6)5,733 (103.2)8.9Public
Anonymous submission
CRF_TRA
89. using public detections 		23.3
53.1
±12.2
53.724.2% 30.7% 27,194234,9912,518 (43.2)4,918 (84.3)1.4Public
Anonymous submission
AReid17
90. online method using public detections 		26.3
51.4
±12.2
53.919.2% 32.3% 30,079241,3642,993 (52.3)6,373 (111.4)33.7Public
Anonymous submission
TrackerAvg RankMOTA IDF1MTMLFPFNID Sw.FragHzDetector
YOONKJ17
91. online method using public detections 		34.3
51.4
±13.5
54.021.2% 37.3% 29,051243,2022,118 (37.2)3,072 (54.0)3.4Public
Anonymous submission
NOTA
92. using public detections 		30.1
51.3
±11.7
54.517.1% 35.4% 20,148252,5312,285 (41.4)5,798 (105.0)17.8Public
SPL-26677-2019
jCC
93. using public detections 		29.4
51.2
±14.5
54.520.9% 37.0% 25,937247,8221,802 (32.1)2,984 (53.2)1.8Public
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.
STCG17
94. using public detections 		28.3
51.1
±12.9
54.520.4% 38.6% 32,258241,9161,702 (29.8)2,483 (43.5)66.9Public
Anonymous submission
eHAF17
95. using public detections 		28.1
51.8
±13.2
54.723.4% 37.9% 33,212236,7721,834 (31.6)2,739 (47.2)0.7Public
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.
overlap
96. using public detections 		24.1
51.5
±13.1
55.623.0% 36.1% 38,322233,2751,860 (31.7)2,935 (50.0)66.9Public
Anonymous submission
DMAN
97. online method using public detections 		38.3
48.2
±12.3
55.719.3% 38.3% 26,218263,6082,194 (41.2)5,378 (100.9)0.3Public
J. Zhu, H. Yang, N. Liu, M. Kim, W. Zhang, M. Yang. Online Multi-Object Tracking with Dual Matching Attention Networks. In ECCV, 2018.
TLMHT
98. using public detections 		36.7
50.6
±12.5
56.517.6% 43.4% 22,213255,0301,407 (25.7)2,079 (37.9)2.6Public
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.
ENFT17
99. using public detections 		23.9
52.8
±13.1
57.123.1% 36.8% 26,754237,9091,667 (28.8)2,557 (44.2)0.5Public
BUAA
SAS_MOT17
100. using public detections 		47.6
44.2
±12.2
57.216.1% 44.3% 29,473283,6111,529 (30.7)2,644 (53.2)4.8Public
A. Maksai, P. Fua. Eliminating Exposure Bias and Metric Mismatch in Multiple Object Tracking. In CVPR, 2019.
TrackerAvg RankMOTA IDF1MTMLFPFNID Sw.FragHzDetector
LSST17O
101. online method using public detections 		33.0
52.7
±13.3
57.917.9% 36.6% 22,512241,9362,167 (37.9)7,443 (130.3)1.8Public
Multi-Object Tracking with Multiple Cues and Switcher-Aware Classification
eTC17
102. using public detections 		27.3
51.9
±12.4
58.123.1% 35.5% 36,164232,7832,288 (38.9)3,071 (52.3)0.7Public
G. Wang, Y. Wang, H. Zhang, R. Gu, J. Hwang. Exploit the Connectivity: Multi-Object Tracking with TrackletNet. In arXiv preprint arXiv:1811.07258, 2018.
CMT
103. using public detections 		24.7
51.8
±12.9
60.719.6% 42.8% 29,528240,9601,217 (21.2)2,008 (35.0)6.5Public
#Submission: TCSVT-02964-2019
IDGA
104. using public detections 		20.7
52.6
±13.4
61.323.6% 40.2% 29,049236,8301,402 (24.2)2,613 (45.0)59.2Public
Anonymous submission
LSST17
105. using public detections 		27.8
54.7
±12.9
62.320.4% 40.1% 26,091228,4341,243 (20.9)3,726 (62.6)1.5Public
Multi-Object Tracking with Multiple Cues and Switcher-Aware Classification
ISDH_HDAv2
106. online method using public detections 		28.3
54.5
±14.5
65.926.4% 32.1% 46,693207,0933,010 (47.6)6,000 (94.8)3.6Public
MM-008988/ IEEE Transactions on Multimedia

Benchmark Statistics

SequencesFramesTrajectoriesBoxes
21177572355564228

Difficulty Analysis

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

MOT17-03-SDP

MOT17-03-SDP

(70.7% MOTA)

MOT17-03-FRCNN

MOT17-03-FRCNN

(57.6% MOTA)

MOT17-06-SDP

MOT17-06-SDP

(47.7% MOTA)

...

...

MOT17-14-DPM

MOT17-14-DPM

(17.7% MOTA)

MOT17-14-FRCNN

MOT17-14-FRCNN

(17.1% 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.



Contact