Head Tracking 21 Results
Click on a measure to sort the table accordingly. See below for a more detailed description.
| Tracker | MOTA | IDF1 | IDEucl | HOTA | MT | ML | FP | FN | Rcll | Prcn | DetPr | DetRe | ID Sw. | Frag | Hz | |
| HT21_FUJITSU 1.  | 82.0 | 72.4 | 71.3 | 51.9 | 1,577 (67.3) | 159 (6.8) | 40,371 | 107,829 | 87.2 | 94.8 | 67.8 | 62.4 | 3,644 (0.0) | 10,799 (0.0) | 572.3 | |
| MiniTrackHT 2. | 80.9 | 72.7 | 73.7 | 52.7 | 1,819 (77.7) | 112 (4.8) | 82,569 | 75,126 | 91.1 | 90.3 | 65.6 | 66.2 | 3,182 (0.0) | 5,959 (0.0) | 9.2 | |
| AIICTHT 3. | 79.7 | 72.9 | 72.5 | 50.7 | 1,635 (69.8) | 170 (7.3) | 61,287 | 106,497 | 87.4 | 92.3 | 64.5 | 61.0 | 2,821 (0.0) | 7,706 (0.0) | 19.1 | |
| Q.He, J.Mei, Z.Ou, AIIC Vision, Midea Group | ||||||||||||||||
| AntTracking 4. | 76.2 | 74.3 | 71.7 | 50.8 | 1,481 (63.2) | 332 (14.2) | 52,879 | 145,374 | 82.7 | 92.9 | 65.3 | 58.2 | 1,729 (0.0) | 5,987 (0.0) | 21.6 | |
| SDTrack_HT 5. | 74.6 | 68.3 | 66.3 | 47.7 | 1,320 (56.4) | 246 (10.5) | 62,538 | 147,930 | 82.4 | 91.7 | 63.8 | 57.3 | 3,185 (0.0) | 6,493 (0.0) | 10.4 | |
| pptracking 6. | 72.6 | 61.8 | 59.7 | 44.6 | 1,282 (54.7) | 234 (10.0) | 71,235 | 154,139 | 81.7 | 90.6 | 62.7 | 56.5 | 5,163 (0.0) | 11,697 (0.0) | 715.4 | |
| https://github.com/PaddlePaddle/PaddleDetection | ||||||||||||||||
| PPMOT_HT 7. | 71.4 | 60.9 | 59.6 | 44.4 | 1,275 (54.4) | 237 (10.1) | 78,101 | 157,762 | 81.3 | 89.8 | 62.4 | 56.5 | 5,296 (0.0) | 12,350 (0.0) | 715.4 | |
| https://github.com/PaddlePaddle/PaddleDetection | ||||||||||||||||
| THT 8. | 70.7 | 68.4 | 63.5 | 47.3 | 1,070 (45.7) | 366 (15.6) | 33,545 | 211,162 | 74.9 | 95.0 | 67.4 | 53.2 | 2,393 (0.0) | 8,204 (0.0) | 11.2 | |
| Q.He, J.Mei, Z.Ou, AIIC Vision, Midea Group | ||||||||||||||||
| FM_OCSORT 9. | 67.9 | 62.9 | 62.1 | 44.1 | 1,269 (54.2) | 221 (9.4) | 102,050 | 164,090 | 80.5 | 86.9 | 60.2 | 55.8 | 4,243 (0.0) | 10,122 (0.0) | 17.7 | |
| J. Cao, X. Weng, R. Khirodkar, J. Pang, K. Kitani. Observation-Centric SORT: Rethinking SORT for Robust Multi-Object Tracking. In , 2022. | ||||||||||||||||
| fairmot_head 10. | 60.8 | 62.8 | 69.9 | 43.0 | 1,057 (45.1) | 174 (7.4) | 118,109 | 198,896 | 76.4 | 84.5 | 58.6 | 53.0 | 12,781 (0.0) | 41,399 (0.0) | 715.4 | |
| Tracker | MOTA | IDF1 | IDEucl | HOTA | MT | ML | FP | FN | Rcll | Prcn | DetPr | DetRe | ID Sw. | Frag | Hz | |
| PHDTT 11. | 60.6 | 47.9 | 52.6 | 36.1 | 1,104 (47.1) | 188 (8.0) | 132,714 | 184,215 | 78.1 | 83.2 | 57.7 | 54.2 | 15,004 (0.0) | 41,556 (0.0) | 21.2 | |
| Vo, Xuan-Thuy, et al. "Pedestrian Head Detection and Tracking via Global Vision Transformer." International Workshop on Frontiers of Computer Vision. Springer, Cham, 2022. | ||||||||||||||||
| HeadHunterT 12. | 57.8 | 53.9 | 54.2 | 36.8 | 747 (31.9) | 465 (19.9) | 51,840 | 299,459 | 64.4 | 91.3 | 60.8 | 42.9 | 4,394 (0.0) | 15,146 (0.0) | 0.4 | |
| R. Sundararaman, C. De Almeida Braga, E. Marchand, J. Pettre. Tracking Pedestrian Heads in Dense Crowd. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021. | ||||||||||||||||
| ToH 13. | 56.7 | 57.4 | 55.8 | 39.0 | 851 (36.3) | 323 (13.8) | 104,580 | 254,298 | 69.8 | 84.9 | 57.8 | 47.5 | 5,551 (0.0) | 23,955 (0.0) | 3.5 | |
| CTv0 14. | 52.4 | 35.8 | 35.2 | 28.0 | 640 (27.3) | 519 (22.2) | 71,037 | 320,820 | 61.9 | 88.0 | 59.2 | 41.6 | 8,844 (0.0) | 12,727 (0.0) | 261.8 | |
| J. Lohn-Jaramillo, L. Ray, R. Granger, E. Bowen. ClusterTracker: An Efficiency-Focused Multiple Object Tracking Method. In , 2022. | ||||||||||||||||
| Sequences | Frames | Trajectories | Boxes |
| 5 | 5723 | 2965 | 1086790 |
Evaluation Measures
Lower is better. Higher is better.| Measure | Better | Perfect | Description |
| MOTA | higher | 100% | Multi-Object Tracking Accuracy (+/- denotes standard deviation across all sequences) [1]. This measure combines three error sources: false positives, missed targets and identity switches. |
| IDF1 | higher | 100% | ID F1 Score [2]. The ratio of correctly identified detections over the average number of ground-truth and computed detections. |
| IDEucl | higher | 100% | IDEucl [3]. Duration correctly tracked in image coordinates (IDEucl = % of length covered in image coordinate),IDEucl,ASC,Computes the ratio of the track length covered by the best possible hypothesis (in terms of image coordinates) to the ground truth trajectory le... |
| HOTA | higher | 100% | Higher Order Tracking Accuracy [4]. Geometric mean of detection accuracy and association accuracy. Averaged across localization thresholds. |
| 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). |
| Rcll | higher | 100% | Ratio of correct detections to total number of GT boxes. |
| Prcn | higher | 100% | Ratio of TP / (TP+FP). |
| DetPr | higher | 100% | Detection Precision [4]. TP /(TP + FP) averaged over localization thresholds. |
| DetRe | higher | 100% | Detection Recall [4]. TP /(TP + FN) averaged over localization thresholds. |
| ID Sw. | lower | 0 | Number of Identity Switches (ID switch ratio = #ID switches / recall) [5]. Please note that we follow the stricter definition of identity switches as described in the reference |
| Frag | lower | 0 | The total number of times a trajectory is fragmented (i.e. interrupted during tracking). |
| Hz | higher | Inf. | Processing speed (in frames per second excluding the detector) on the benchmark. The frequency is provided by the authors and not officially evaluated by the MOTChallenge. |
Legend
| Symbol | Description |
|
This is an online (causal) method, i.e. the solution is immediately available with each incoming frame and cannot be changed at any later time. |
 |
This method used the provided detection set as input. |
 |
This method used a private detection set as input. |
 |
This entry has been submitted or updated less than a week ago. |
References:
| [1] | Evaluating Multiple Object Tracking Performance: The CLEAR MOT Metrics. Image and Video Processing, 2008(1):1-10, 2008. |
| [2] | Performance Measures and a Data Set for Multi-Target, Multi-Camera Tracking. In ECCV workshop on Benchmarking Multi-Target Tracking, 2016. |
| [3] | Tracking Pedestrian Heads in Dense Crowd. In Conference on Computer Vision and Pattern Recognition (CVPR), 2021. |
| [4] | HOTA: A Higher Order Metric for Evaluating Multi-Object Tracking. International Journal of Computer Vision, 2020. |
| [5] | 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. |
