LPC_MOT: Learning a Proposal Classifier for Multiple Object Tracking
MOT16-01
Short name:
LPC_MOT
Detector:
Public
Description:
This approach is devised by Tsinghua University and AIBEE Inc.
Reference:
P. Dai, R. Weng, W. Choi, C. Zhang, Z. He, W. Ding. Learning a Proposal Classifier for Multiple Object tracking. In CVPR (Accepted), 2021.
Last submitted:
January 22, 2021 (1 year ago)
Published:
January 22, 2021 at 12:57:20 CET
Submissions:
1
Project page / code:
Open source:
Yes
Hardware:
GTX 1080 TI 20 cores
Runtime:
4.3 Hz
Benchmark performance:
| Sequence | MOTA | IDF1 | HOTA | MT | ML | FP | FN | Rcll | Prcn | AssA | DetA | AssRe | AssPr | DetRe | DetPr | LocA | FAF | ID Sw. | Frag |
| MOT16 | 58.8 | 67.6 | 51.7 | 207 (27.3) | 266 (35.0) | 6,167 | 68,432 | 62.5 | 94.9 | 56.4 | 47.6 | 60.2 | 80.1 | 50.5 | 76.7 | 81.3 | 1.0 | 435 (7.0) | 628 (10.1) |
Detailed performance:
| Sequence | MOTA | IDF1 | HOTA | MT | ML | FP | FN | Rcll | Prcn | AssA | DetA | AssRe | AssPr | DetRe | DetPr | LocA | FAF | ID Sw. | Frag |
| MOT16-01 | 47.6 | 55.1 | 45.2 | 8 | 10 | 64 | 3,271 | 48.9 | 98.0 | 54.5 | 37.5 | 56.9 | 80.8 | 38.8 | 77.8 | 80.7 | 0.1 | 14 | 16 |
| MOT16-03 | 69.8 | 76.3 | 57.4 | 71 | 19 | 3,238 | 28,187 | 73.0 | 95.9 | 60.0 | 55.2 | 62.7 | 81.4 | 58.7 | 77.1 | 81.2 | 2.2 | 106 | 161 |
| MOT16-06 | 55.1 | 57.8 | 46.2 | 70 | 86 | 741 | 4,373 | 62.1 | 90.6 | 45.8 | 46.9 | 62.6 | 65.7 | 51.0 | 74.4 | 81.8 | 0.6 | 65 | 88 |
| MOT16-07 | 43.8 | 54.5 | 41.5 | 8 | 11 | 1,325 | 7,761 | 52.5 | 86.6 | 44.1 | 39.5 | 48.1 | 75.4 | 43.0 | 71.1 | 79.7 | 2.7 | 90 | 155 |
| MOT16-08 | 42.9 | 54.8 | 45.8 | 17 | 24 | 205 | 9,313 | 44.4 | 97.3 | 58.4 | 36.1 | 61.7 | 84.0 | 37.3 | 81.8 | 84.0 | 0.3 | 32 | 48 |
| MOT16-12 | 49.2 | 62.3 | 48.8 | 20 | 38 | 231 | 3,961 | 52.2 | 94.9 | 57.1 | 41.8 | 61.0 | 81.4 | 44.0 | 79.9 | 84.0 | 0.3 | 22 | 28 |
| MOT16-14 | 34.9 | 45.4 | 32.6 | 13 | 78 | 363 | 11,566 | 37.4 | 95.0 | 37.4 | 28.6 | 39.3 | 78.3 | 29.7 | 75.4 | 80.1 | 0.5 | 106 | 132 |
Raw data:
