Lif_T: Lifted Disjoint Paths with Application in Multiple Object Tracking
MOT17-12-SDP
Benchmark:
MOT17 |
Short name:
Lif_T
Detector:
Public
Description:
n/a
Reference:
A. Hornakova, R. Henschel, B. Rosenhahn, P. Swoboda. Lifted Disjoint Paths with Application in Multiple Object Tracking. In ICML, 2020.
Last submitted:
January 29, 2020 (6 months ago)
Published:
June 26, 2020 at 09:34:41 CET
Submissions:
2
Project page / code:
Open source:
Yes
Hardware:
2GHZ. 6 Cores
Runtime:
0.5 Hz
Benchmark performance:
| Sequence | MOTA | IDF1 | MOTP | MT | ML | FP | FN | Recall | Precision | FAF | ID Sw. | Frag |
| MOT17 | 60.5 | 65.6 | 78.3 | 637 (27.0) | 791 (33.6) | 14,966 | 206,619 | 63.4 | 96.0 | 0.8 | 1,189 (18.8) | 3,476 (54.8) |
Detailed performance:
| Sequence | MOTA | IDF1 | MOTP | MT | ML | FP | FN | Recall | Precision | FAF | ID Sw. | Frag |
| MOT17-01-DPM | 48.3 | 58.1 | 76.8 | 8 | 11 | 68 | 3,258 | 49.5 | 97.9 | 0.2 | 10 | 19 |
| MOT17-01-FRCNN | 47.7 | 58.1 | 77.3 | 8 | 10 | 246 | 3,119 | 51.6 | 93.1 | 0.5 | 7 | 24 |
| MOT17-01-SDP | 47.8 | 57.8 | 76.9 | 9 | 10 | 346 | 3,008 | 53.4 | 90.9 | 0.8 | 10 | 31 |
| MOT17-03-DPM | 73.3 | 70.1 | 77.9 | 82 | 17 | 3,560 | 24,276 | 76.8 | 95.8 | 2.4 | 160 | 256 |
| MOT17-03-FRCNN | 72.2 | 71.8 | 78.2 | 71 | 17 | 2,664 | 26,277 | 74.9 | 96.7 | 1.8 | 124 | 250 |
| MOT17-03-SDP | 78.2 | 77.3 | 77.5 | 92 | 13 | 3,778 | 18,879 | 82.0 | 95.8 | 2.5 | 132 | 323 |
| MOT17-06-DPM | 58.1 | 64.7 | 79.8 | 61 | 77 | 178 | 4,728 | 59.9 | 97.5 | 0.1 | 28 | 155 |
| MOT17-06-FRCNN | 60.4 | 63.7 | 80.1 | 68 | 61 | 279 | 4,358 | 63.0 | 96.4 | 0.2 | 32 | 207 |
| MOT17-06-SDP | 60.3 | 65.1 | 79.9 | 67 | 64 | 305 | 4,345 | 63.1 | 96.1 | 0.3 | 33 | 217 |
| MOT17-07-DPM | 44.4 | 52.3 | 78.6 | 7 | 21 | 155 | 9,176 | 45.7 | 98.0 | 0.3 | 60 | 209 |
| MOT17-07-FRCNN | 44.0 | 54.9 | 78.6 | 8 | 20 | 279 | 9,110 | 46.1 | 96.5 | 0.6 | 63 | 227 |
| MOT17-07-SDP | 45.8 | 55.0 | 78.5 | 8 | 18 | 285 | 8,793 | 47.9 | 96.6 | 0.6 | 71 | 280 |
| MOT17-08-DPM | 34.7 | 47.4 | 81.8 | 18 | 37 | 254 | 13,507 | 36.1 | 96.8 | 0.4 | 32 | 44 |
| MOT17-08-FRCNN | 31.9 | 43.3 | 81.7 | 17 | 37 | 383 | 13,973 | 33.9 | 94.9 | 0.6 | 35 | 59 |
| MOT17-08-SDP | 34.8 | 47.7 | 81.1 | 18 | 34 | 429 | 13,288 | 37.1 | 94.8 | 0.7 | 48 | 69 |
| MOT17-12-DPM | 48.3 | 62.3 | 81.8 | 18 | 41 | 35 | 4,437 | 48.8 | 99.2 | 0.0 | 11 | 52 |
| MOT17-12-FRCNN | 47.3 | 58.0 | 81.8 | 16 | 43 | 37 | 4,521 | 47.8 | 99.1 | 0.0 | 11 | 34 |
| MOT17-12-SDP | 47.3 | 60.7 | 81.5 | 18 | 42 | 158 | 4,394 | 49.3 | 96.4 | 0.2 | 14 | 53 |
| MOT17-14-DPM | 36.1 | 48.8 | 77.1 | 12 | 77 | 268 | 11,449 | 38.1 | 96.3 | 0.4 | 91 | 239 |
| MOT17-14-FRCNN | 36.2 | 49.0 | 75.9 | 16 | 72 | 629 | 11,061 | 40.2 | 92.2 | 0.8 | 108 | 358 |
| MOT17-14-SDP | 38.3 | 51.4 | 76.1 | 15 | 69 | 630 | 10,662 | 42.3 | 92.5 | 0.8 | 109 | 370 |
Raw data:
