LP_SSVM: Linear Programming Learned With Structured SVM
Video not available.
Rendering of new sequences is currently deactivated due to heavy load.
Rendering of new sequences is currently deactivated due to heavy load.
Rendering of new sequences is currently deactivated due to heavy load.
Benchmark:
MOT15 |
Short name:
LP_SSVM
Detector:
Public
Description:
Min-cost network flow model with pairwise interactions; network cost is learned using structured SVM.
Reference:
S. Wang, C. Fowlkes. Learning Optimal Parameters for Multi-target Tracking with Contextual Interactions. In International Journal of Computer Vision, 2016.
Last submitted:
April 20, 2015 (9 years ago)
Published:
April 21, 2015 at 04:22:07 CET
Submissions:
1
Project page / code:
n/a
Open source:
No
Hardware:
2.4GHz, 1 Core
Runtime:
41.3 Hz
Benchmark performance:
| Sequence | MOTA | IDF1 | HOTA | MT | ML | FP | FN | Rcll | Prcn | AssA | DetA | AssRe | AssPr | DetRe | DetPr | LocA | FAF | ID Sw. | Frag |
| MOT15 | 25.2 | 34.0 | 26.3 | 42 (5.8) | 382 (53.0) | 8,369 | 36,932 | 39.9 | 74.5 | 26.3 | 26.7 | 29.0 | 66.2 | 30.4 | 56.9 | 75.6 | 1.4 | 646 (16.2) | 849 (21.3) |
Detailed performance:
| Sequence | MOTA | IDF1 | HOTA | MT | ML | FP | FN | Rcll | Prcn | AssA | DetA | AssRe | AssPr | DetRe | DetPr | LocA | FAF | ID Sw. | Frag |
| ADL-Rundle-1 | 14.0 | 38.3 | 30.2 | 7 | 7 | 3,507 | 4,423 | 52.5 | 58.2 | 32.2 | 28.9 | 35.0 | 69.3 | 40.0 | 44.4 | 75.9 | 7.0 | 69 | 92 |
| ADL-Rundle-3 | 28.0 | 36.9 | 28.5 | 4 | 10 | 1,855 | 5,388 | 47.0 | 72.0 | 27.1 | 30.5 | 29.0 | 69.0 | 36.2 | 55.4 | 76.2 | 3.0 | 81 | 83 |
| AVG-TownCentre | 14.7 | 24.3 | 19.7 | 6 | 139 | 459 | 5,515 | 22.8 | 78.1 | 23.4 | 16.9 | 28.7 | 53.6 | 17.9 | 61.0 | 73.7 | 1.0 | 123 | 141 |
| ETH-Crossing | 24.9 | 37.5 | 26.0 | 1 | 17 | 10 | 741 | 26.1 | 96.3 | 34.3 | 19.9 | 35.3 | 80.6 | 20.3 | 75.0 | 79.2 | 0.0 | 2 | 2 |
| ETH-Jelmoli | 39.5 | 52.4 | 35.5 | 8 | 16 | 224 | 1,293 | 49.0 | 84.7 | 36.7 | 34.6 | 41.0 | 69.8 | 38.1 | 65.9 | 77.9 | 0.5 | 17 | 29 |
| ETH-Linthescher | 15.6 | 23.5 | 21.0 | 5 | 157 | 41 | 7,483 | 16.2 | 97.2 | 35.7 | 12.4 | 37.9 | 76.0 | 12.5 | 75.3 | 79.4 | 0.0 | 11 | 15 |
| KITTI-16 | 39.2 | 41.7 | 27.3 | 0 | 2 | 90 | 924 | 45.7 | 89.6 | 23.3 | 32.3 | 26.9 | 57.9 | 34.7 | 68.0 | 77.4 | 0.4 | 20 | 29 |
| KITTI-19 | 28.2 | 39.6 | 27.1 | 4 | 18 | 810 | 2,955 | 44.7 | 74.7 | 25.3 | 29.6 | 29.7 | 58.7 | 33.3 | 55.6 | 71.9 | 0.8 | 70 | 162 |
| PETS09-S2L2 | 41.5 | 27.9 | 21.1 | 3 | 7 | 629 | 4,803 | 50.2 | 88.5 | 12.9 | 35.2 | 13.9 | 59.8 | 37.6 | 66.2 | 74.6 | 1.4 | 212 | 249 |
| TUD-Crossing | 60.0 | 55.0 | 38.3 | 4 | 2 | 48 | 375 | 66.0 | 93.8 | 31.8 | 46.2 | 38.6 | 57.3 | 49.8 | 70.8 | 77.7 | 0.2 | 18 | 20 |
| Venice-1 | 17.8 | 31.9 | 25.0 | 0 | 7 | 696 | 3,032 | 33.6 | 68.7 | 28.0 | 22.5 | 28.9 | 76.9 | 26.1 | 53.4 | 76.6 | 1.5 | 23 | 27 |
Raw data:
