2D MOT 2015 Results

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


Showing only entries that use public detections!

TrackerAvg RankMOTAIDF1MTMLFPFNID Sw.FragHzDetector
KCF
1. online method using public detections
17.3
38.9
±14.5
44.516.6% 31.5% 7,32129,501720 (13.9)1,440 (27.7)0.3Public
Paper ID 2988
AP_HWDPL_p
2. online method using public detections
14.8
38.5
±9.9
47.18.7% 37.4% 4,00533,203586 (12.8)1,263 (27.5)6.7Public
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.
AMIR15
3. online method using public detections
17.7
37.6
±12.5
46.015.8% 26.8% 7,93329,3971,026 (19.7)2,024 (38.8)1.9Public
A. Sadeghian, A. Alahi, S. Savarese. Tracking The Untrackable: Learning To Track Multiple Cues with Long-Term Dependencies. In ICCV, 2017.
TPM
4. using public detections
20.6
36.2
±12.8
43.615.4% 42.6% 5,65033,102420 (9.1)928 (20.1)0.8Public
Anonymous submission
JointMC
5. using public detections
17.8
35.6
±18.9
45.123.2% 39.3% 10,58028,508457 (8.5)969 (18.1)0.6Public
M. Keuper, S. Tang, Z. Yu, B. Andres, T. Brox, B. Schiele. A Multi-cut Formulation for Joint Segmentation and Tracking of Multiple Objects. In CoRR, 2016.
mLK
6. online method using public detections
17.7
35.1
±12.9
47.512.3% 38.3% 5,67833,815383 (8.5)1,175 (26.1)1.0Public
Yuan Zhang, Di Xie and Shiliang Pu (Hikvision Research Institute)
HybridDAT
7. online method using public detections
17.3
35.0
±15.0
47.711.4% 42.2% 8,45531,140358 (7.3)1,267 (25.7)4.6Public
M. Yang, Y. Jia. A Hybrid Data Association Framework for Robust Online Multi-Object Tracking. In IEEE Transactions on Image Processing, 2016.
AM
8. online method using public detections
18.3
34.3
±13.7
48.311.4% 43.4% 5,15434,848348 (8.0)1,463 (33.8)0.5Public
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.
TSMLCDEnew
9. using public detections
20.0
34.3
±13.1
44.114.0% 39.4% 7,86931,908618 (12.9)959 (20.0)6.5Public
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.
QuadMOT
10. using public detections
22.3
33.8
±14.8
40.412.9% 36.9% 7,89832,061703 (14.7)1,430 (29.9)3.7Public
J. Son, M. Baek, M. Cho, B. Han. Multi-Object Tracking with Quadruplet Convolutional Neural Networks. In CVPR, 2017.
TrackerAvg RankMOTAIDF1MTMLFPFNID Sw.FragHzDetector
NOMT
11. using public detections
17.6
33.7
±16.2
44.612.2% 44.0% 7,76232,547442 (9.4)823 (17.5)11.5Public
W. Choi. Near-Online Multi-target Tracking with Aggregated Local Flow Descriptor. In ICCV, 2015.
DCCRF
12. online method using public detections
25.3
33.6
±11.0
39.110.4% 37.6% 5,91734,002866 (19.4)1,566 (35.1)0.1Public
H. Zhou, W. Ouyang, J. Cheng, X. Wang, and H. Li, Deep Continuous Conditional Random Fields with Asymmetric Inter-object Constraints for Online Multi-object Tracking, IEEE Transactions on Circuits and Systems for Video Technology.
TDAM
13. online method using public detections
21.9
33.0
±9.8
46.113.3% 39.1% 10,06430,617464 (9.2)1,506 (30.0)5.9Public
M. Yang, Y. Jia. Temporal dynamic appearance modeling for online multi-person tracking. In Computer Vision and Image Understanding, 2016.
CDA_DDALpb
14. online method using public detections
22.8
32.8
±10.6
38.89.7% 42.2% 4,98335,690614 (14.7)1,583 (37.8)2.3Public
S. Bae and K. Yoon, Confidence-Based Data Association and Discriminative Deep Appearance Learning for Robust Online Multi-Object Tracking , In IEEE TPAMI, 2017.
MHT_DAM
15. using public detections
19.9
32.4
±15.6
45.316.0% 43.8% 9,06432,060435 (9.1)826 (17.3)0.7Public
C. Kim, F. Li, A. Ciptadi, J. Rehg. Multiple Hypothesis Tracking Revisited. In ICCV, 2015.
LFNF
16. using public detections
26.3
31.6
±12.3
33.19.6% 41.7% 5,94335,095961 (22.4)1,106 (25.8)4.0Public
Sheng H, Hao L, Chen J, et al. Robust Local Effective Matching Model for Multi-Target Tracking. In PCM, 2017
Q_cf
17. online method using public detections
24.1
31.6
±13.5
43.816.0% 34.7% 9,71131,579742 (15.3)1,505 (31.0)1.0Public
Anonymous submission
cf_mdp
18. online method using public detections
16.0
31.4
±14.7
43.716.1% 34.3% 10,11831,284718 (14.6)1,469 (29.9)450.4Public
Anonymous submission
CF_MCMC
19. using public detections
26.8
31.4
±11.3
36.410.3% 40.9% 8,79832,541814 (17.3)1,711 (36.4)3.2Public
Anonymous submission
SNM
20. online method using public detections
26.9
31.3
±16.5
38.212.6% 35.4% 8,90332,393926 (19.6)2,382 (50.4)14.8Public
Anonymous submission
TrackerAvg RankMOTAIDF1MTMLFPFNID Sw.FragHzDetector
PHD_GSDL
21. online method using public detections
29.4
30.5
±14.9
38.87.6% 41.2% 6,53435,284879 (20.6)2,208 (51.9)8.2Public
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.
AMT
22. online method using public detections
17.7
30.4
±13.6
43.915.0% 35.6% 10,39131,641725 (14.9)1,641 (33.8)296.0Public
Anonymous submission
MDP
23. online method using public detections
24.9
30.3
±14.6
44.713.0% 38.4% 9,71732,422680 (14.4)1,500 (31.8)1.1Public
Y. Xiang, A. Alahi, S. Savarese. Learning to Track: Online Multi-Object Tracking by Decision Making. In International Conference on Computer Vision (ICCV), 2015.
ARMS
24. using public detections
26.3
30.1
±13.4
43.311.8% 33.8% 9,11732,0421,785 (37.3)2,483 (51.9)1.7Public
Anonymous submission
MCF_PHD
25. using public detections
23.8
29.9
±20.0
38.211.9% 44.0% 8,89233,529656 (14.4)989 (21.8)12.2Public
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.
CNNTCM
26. using public detections
25.3
29.6
±13.9
36.811.2% 44.0% 7,78634,733712 (16.4)943 (21.7)1.7Public
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.
MAR
27. using public detections
29.4
29.6
±11.6
42.811.7% 34.3% 9,36432,1311,783 (37.4)2,439 (51.1)1.2Public
Anonymous submission
RSCNN
28. using public detections
26.4
29.5
±23.9
37.012.9% 36.3% 11,86630,474976 (19.4)1,176 (23.3)4.0Public
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.
Stitiching
29. using public detections
30.0
29.3
±25.1
42.611.8% 34.1% 9,55532,0181,877 (39.2)2,428 (50.7)4.8Public
Anonymous submission
Otakudj
30. online method using public detections
30.8
29.2
±13.9
42.212.3% 33.8% 9,55832,0081,924 (40.2)2,463 (51.4)0.6Public
Anonymous submission
TrackerAvg RankMOTAIDF1MTMLFPFNID Sw.FragHzDetector
TBSS15
31. online method using public detections
29.7
29.2
±12.5
37.26.8% 43.8% 6,06836,779649 (16.2)1,508 (37.6)11.5Public
Anonymous submission
SCEA
32. online method using public detections
28.6
29.1
±12.2
37.28.9% 47.3% 6,06036,912604 (15.1)1,182 (29.6)6.8Public
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.
SiameseCNN
33. using public detections
28.6
29.0
±15.1
34.38.5% 48.4% 5,16037,798639 (16.6)1,316 (34.2)52.8Public
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.
ARM
34. using public detections
27.4
29.0
±12.8
40.610.0% 36.3% 7,86633,8211,948 (43.3)2,711 (60.3)9.6Public
Anonymous submission
TBX
35. using public detections
36.6
27.5
±13.3
33.810.4% 45.8% 7,96835,810759 (18.2)1,528 (36.6)0.1Public
R. Henschel, L. Leal-Taixé, B. Rosenhahn, K. Schindler. Tracking with multi-level features. In arXiv:1607.07304, 2016.
oICF
36. online method using public detections
31.8
27.1
±14.9
40.56.4% 48.7% 7,59436,757454 (11.3)1,660 (41.3)1.4Public
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.
RMM
37. using public detections
36.8
26.6
±12.3
31.25.1% 44.8% 6,70837,2811,088 (27.7)1,957 (49.8)9.4Public
Anonymous submission
TO
38. using public detections
32.4
25.7
±13.5
32.74.3% 57.4% 4,77940,511383 (11.2)600 (17.6)5.0Public
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.
LP_SSVM
39. using public detections
29.3
25.2
±13.7
34.05.8% 53.0% 8,36936,932646 (16.2)849 (21.3)41.3Public
S. Wang, C. Fowlkes. Learning Optimal Parameters for Multi-target Tracking with Contextual Interactions. In International Journal of Computer Vision, 2016.
ELP
40. using public detections
34.7
25.0
±10.8
26.27.5% 43.8% 7,34537,3441,396 (35.6)1,804 (46.0)5.7Public
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.
TrackerAvg RankMOTAIDF1MTMLFPFNID Sw.FragHzDetector
AdTobKF
41. online method using public detections
28.3
24.8
±12.1
34.54.0% 52.0% 6,20139,321666 (18.5)1,300 (36.1)206.5Public
Anonymous submission
LINF1
42. using public detections
28.5
24.5
±15.4
34.85.5% 64.6% 5,86440,207298 (8.6)744 (21.5)7.5Public
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.
TENSOR
43. using public detections
37.0
24.3
±13.2
24.15.5% 46.6% 6,64438,5821,271 (34.2)1,304 (35.1)24.0Public
Anonymous submission
TFMOT
44. online method using public detections
32.7
23.8
±11.9
32.34.9% 62.0% 4,53341,873404 (12.7)792 (24.9)11.3Public
M. Abhijeet Boragule. Joint Cost Minimization for Multi-Object Tracking. In 2017 IEEE International Conference on Advanced Vide and Signale Based Surveillance, 2017.
JPDA_m
45. using public detections
28.7
23.8
±15.1
33.85.0% 58.1% 6,37340,084365 (10.5)869 (25.0)32.6Public
H. Rezatofighi, A. Milan, Z. Zhang, Q. Shi, A. Dick, I. Reid. Joint Probabilistic Data Association Revisited. In ICCV, 2015.
HAM_ACT
46. online method using public detections
27.7
23.7
±13.0
35.94.2% 65.9% 5,19141,462237 (7.3)671 (20.6)17.7Public
Anonymous submission
CNNEMBED
47. online method using public detections
24.8
23.6
±14.9
41.012.9% 44.1% 14,04732,398496 (10.5)829 (17.5)1.9Public
Anonymous submission
MotiCon
48. using public detections
39.6
23.1
±16.4
29.44.7% 52.0% 10,40435,8441,018 (24.4)1,061 (25.5)1.4Public
L. Leal-Taixé, M. Fenzi, A. Kuznetsova, B. Rosenhahn, S. Savarese. Learning an image-based motion context for multiple people tracking. In CVPR, 2014.
SegTrack
49. using public detections
38.6
22.5
±15.2
31.55.8% 63.9% 7,89039,020697 (19.1)737 (20.2)0.2Public
A. Milan, L. Leal-Taixé, K. Schindler, I. Reid. Joint Tracking and Segmentation of Multiple Targets. In CVPR, 2015.
EAMTTpub
50. online method using public detections
34.8
22.3
±14.2
32.85.4% 52.7% 7,92438,982833 (22.8)1,485 (40.6)12.2Public
R. Sanchez-Matilla, F. Poiesi, A. Cavallaro "Multi-target tracking with strong and weak detections" in BMTT ECCVw 2016
TrackerAvg RankMOTAIDF1MTMLFPFNID Sw.FragHzDetector
CppSORT
51. online method using public detections
36.9
21.7
±11.8
26.83.7% 49.1% 8,42238,4541,231 (32.9)2,005 (53.6)1,112.1Public
S. Murray. Real-Time Multiple Object Tracking - A Study on the Importance of Speed. In arXiv preprint arXiv:1709.03572, 2017.
OMT_DFH
52. online method using public detections
30.5
21.2
±17.2
37.37.1% 46.5% 13,21834,657563 (12.9)1,255 (28.8)28.6Public
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.
MTSTracker
53. online method using public detections
34.8
20.6
±18.2
31.99.0% 36.9% 15,16132,2121,387 (29.2)2,357 (49.5)19.3Public
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.
TC_SIAMESE
54. online method using public detections
35.2
20.2
±13.9
32.62.6% 67.5% 6,12742,596294 (9.6)825 (26.9)11.0Public
Anonymous submission
LP2D
55. using public detections
37.1
19.8
±14.2
0.06.7% 41.2% 11,58036,0451,649 (39.9)1,712 (41.4)112.1Public
MOT baseline: Linear programming on 2D image coordinates.
DCO_X
56. using public detections
36.8
19.6
±14.1
31.55.1% 54.9% 10,65238,232521 (13.8)819 (21.7)0.3Public
A. Milan, K. Schindler, S. Roth. Multi-Target Tracking by Discrete-Continuous Energy Minimization. In IEEE PAMI, 2016.
CEM
57. using public detections
36.1
19.3
±17.5
0.08.5% 46.5% 14,18034,591813 (18.6)1,023 (23.4)1.1Public
A. Milan, S. Roth, K. Schindler. Continuous Energy Minimization for Multitarget Tracking. In IEEE TPAMI, 2014.
RNN_LSTM
58. online method using public detections
40.8
19.0
±15.2
17.15.5% 45.6% 11,57836,7061,490 (37.0)2,081 (51.7)165.2Public
A. Milan, S. Rezatofighi, A. Dick, I. Reid, K. Schindler. Online Multi-Target Tracking using Recurrent Neural Networks. In AAAI, 2017.
RAM
59. using public detections
35.1
18.8
±12.4
31.112.6% 33.8% 14,38431,6953,814 (78.8)2,458 (50.8)8.2Public
Anonymous submission
RMOT
60. online method using public detections
39.2
18.6
±17.5
32.65.3% 53.3% 12,47336,835684 (17.1)1,282 (32.0)7.9Public
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.
TrackerAvg RankMOTAIDF1MTMLFPFNID Sw.FragHzDetector
TSDA_OAL
61. online method using public detections
35.7
18.6
±17.6
36.19.4% 42.3% 16,35032,853806 (17.3)1,544 (33.2)19.7Public
H. Ko. Online multi-person tracking with two-stage data association and online appearance model learning. In IET Computer Vision, 2017.
GMPHD
62. online method using public detections
34.8
18.5
±12.7
28.43.9% 55.3% 7,86441,766459 (14.3)1,266 (39.5)19.8Public
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.
SMOT
63. using public detections
47.8
18.2
±10.3
0.02.8% 54.8% 8,78040,3101,148 (33.4)2,132 (62.0)2.7Public
C. Dicle, O. Camps, M. Sznaier. The Way They Move: Tracking Targets with Similar Appearance. In ICCV, 2013.
ALExTRAC
64. using public detections
42.5
17.0
±12.1
17.33.9% 52.4% 9,23339,9331,859 (53.1)1,872 (53.5)3.7Public
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
65. using public detections
47.7
15.9
±17.6
0.06.4% 47.9% 14,94334,7771,939 (44.7)1,963 (45.2)0.7Public
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.
GSCR
66. online method using public detections
35.3
15.8
±10.5
27.91.8% 61.0% 7,59743,633514 (17.7)1,010 (34.8)28.1Public
L. Fagot-Bouquet, R. Audigier, Y. Dhome, F. Lerasle. Online multi-person tracking based on global sparse collaborative representations. In ICIP, 2015.
TC_ODAL
67. online method using public detections
47.1
15.1
±15.0
0.03.2% 55.8% 12,97038,538637 (17.1)1,716 (46.0)1.7Public
S. Bae, K. Yoon. Robust Online Multi-Object Tracking based on Tracklet Confidence and Online Discriminative Appearance Learning. In CVPR, 2014.
DP_NMS
68. using public detections
38.3
14.5
±14.5
19.76.0% 40.8% 13,17134,8144,537 (104.7)3,090 (71.3)444.8Public
H. Pirsiavash, D. Ramanan, C. Fowlkes. Globally-Optimal Greedy Algorithms for Tracking a Variable Number of Objects. In CVPR, 2011.
LDCT
69. online method using public detections
35.7
4.7
±41.3
16.811.4% 32.5% 14,06632,15612,348 (259.1)2,918 (61.2)20.7Public
F. Solera, S. Calderara, R. Cucchiara. Learning to Divide and Conquer for Online Multi-Target Tracking. In ICCV, 2015
Dr_lzy
70. using public detections
32.3
-15.8
±21.6
24.116.6% 28.7% 38,04528,0295,079 (93.4)2,719 (50.0)9.6Public
Anonymous submission

Benchmark Statistics

SequencesFramesTrajectoriesBoxes
11578372161440

Difficulty Analysis

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

TUD-Crossing

TUD-Crossing

(62.1% MOTA)

PETS09-S2L2

PETS09-S2L2

(40.6% MOTA)

ETH-Jelmoli

ETH-Jelmoli

(38.1% MOTA)

...

...

Venice-1

Venice-1

(20.9% MOTA)

ADL-Rundle-1

ADL-Rundle-1

(16.2% 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.