STRN: Spatial-Temporal Relation Networks for Multi-Object Tracking
TUD-Crossing
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.
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:
STRN
Detector:
Public
Description:
Spatial-Temporal Relation Networks for Multi-Object Tracking, ICCV2019,
https://arxiv.org/abs/1904.11489
Reference:
J. Xu, Y. Cao, Z. Zhang, H. Hu. Spatial-Temporal Relation Networks for Multi-Object Tracking. In ICCV, 2019.
Last submitted:
November 04, 2019 (5 years ago)
Published:
November 04, 2019 at 04:46:08 CET
Submissions:
1
Project page / code:
n/a
Open source:
Yes
Hardware:
Titan XP
Runtime:
13.8 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 | 38.1 | 46.6 | 33.5 | 83 (11.5) | 241 (33.4) | 5,451 | 31,571 | 48.6 | 84.6 | 33.5 | 33.8 | 38.5 | 58.7 | 37.1 | 64.5 | 76.0 | 0.9 | 1,033 (21.2) | 2,665 (54.8) |
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 | 40.3 | 53.6 | 39.1 | 6 | 6 | 720 | 4,756 | 48.9 | 86.3 | 44.0 | 34.8 | 47.9 | 66.0 | 38.0 | 67.1 | 77.6 | 1.4 | 76 | 308 |
| ADL-Rundle-3 | 36.5 | 47.7 | 33.6 | 6 | 9 | 1,461 | 4,916 | 51.6 | 78.2 | 33.7 | 33.8 | 37.1 | 66.4 | 38.9 | 58.9 | 75.5 | 2.3 | 74 | 202 |
| AVG-TownCentre | 27.9 | 42.0 | 31.4 | 29 | 60 | 1,165 | 3,763 | 47.4 | 74.4 | 31.4 | 31.9 | 36.7 | 61.6 | 36.4 | 57.2 | 73.4 | 2.6 | 225 | 567 |
| ETH-Crossing | 32.1 | 47.2 | 30.4 | 1 | 14 | 27 | 651 | 35.1 | 92.9 | 36.0 | 25.8 | 38.6 | 69.5 | 26.8 | 70.9 | 77.3 | 0.1 | 3 | 22 |
| ETH-Jelmoli | 51.7 | 58.0 | 39.6 | 9 | 16 | 108 | 1,105 | 56.4 | 93.0 | 39.9 | 39.3 | 47.0 | 61.6 | 42.2 | 69.6 | 77.7 | 0.2 | 13 | 77 |
| ETH-Linthescher | 33.8 | 45.8 | 31.8 | 15 | 110 | 249 | 5,607 | 37.2 | 93.0 | 38.0 | 26.8 | 43.7 | 66.2 | 27.9 | 69.9 | 76.6 | 0.2 | 52 | 170 |
| KITTI-16 | 38.1 | 51.4 | 34.1 | 0 | 1 | 127 | 892 | 47.6 | 86.4 | 34.7 | 33.5 | 38.7 | 58.6 | 36.1 | 65.5 | 74.8 | 0.6 | 34 | 125 |
| KITTI-19 | 26.8 | 49.8 | 33.2 | 4 | 16 | 1,033 | 2,797 | 47.7 | 71.1 | 34.3 | 32.5 | 39.4 | 57.5 | 37.0 | 55.2 | 71.4 | 1.0 | 79 | 374 |
| PETS09-S2L2 | 47.9 | 32.9 | 24.6 | 4 | 4 | 374 | 4,197 | 56.5 | 93.6 | 15.2 | 39.9 | 20.5 | 32.9 | 42.6 | 70.6 | 77.2 | 0.9 | 450 | 643 |
| TUD-Crossing | 68.0 | 64.1 | 44.8 | 6 | 0 | 18 | 324 | 70.6 | 97.7 | 40.8 | 49.2 | 52.3 | 58.3 | 51.9 | 71.9 | 76.8 | 0.1 | 11 | 62 |
| Venice-1 | 39.8 | 51.7 | 36.6 | 3 | 5 | 169 | 2,563 | 43.8 | 92.2 | 41.1 | 32.7 | 46.5 | 65.9 | 34.5 | 72.7 | 79.5 | 0.4 | 16 | 115 |
Raw data:
