RSCNN: multi Correlation filter tracking using hierarchy of convolution features + minimum-cost network flow.
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.
Benchmark:
MOT15 |
Short name:
RSCNN
Detector:
Public
Description:
n/a
Reference:
Heba Mahgoub, Khaled Mostafa, Khaled T. Wassif and Ibrahim Farag, “Multi-Target Tracking Using Hierarchical Convolutional Features and Motion Cues” International Journal of Advanced Computer Science and Applications(IJACSA), 8(11), 2017.
Last submitted:
October 07, 2017 (7 years ago)
Published:
July 08, 2017 at 23:21:06 CET
Submissions:
4
Project page / code:
n/a
Open source:
No
Hardware:
Runtime:
4.0 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 | 29.5 | 37.0 | 29.6 | 93 (12.9) | 262 (36.3) | 11,866 | 30,474 | 50.4 | 72.3 | 27.2 | 32.6 | 30.0 | 67.1 | 39.2 | 56.2 | 76.7 | 2.1 | 976 (19.4) | 1,176 (23.3) |
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 | 20.1 | 33.1 | 28.8 | 8 | 3 | 3,457 | 3,828 | 58.9 | 61.3 | 25.4 | 33.3 | 26.9 | 72.6 | 45.7 | 47.6 | 76.3 | 6.9 | 150 | 152 |
| ADL-Rundle-3 | 36.0 | 39.1 | 30.8 | 7 | 8 | 1,250 | 5,171 | 49.1 | 80.0 | 27.0 | 35.2 | 28.3 | 75.2 | 40.0 | 65.1 | 80.1 | 2.0 | 86 | 81 |
| AVG-TownCentre | 28.6 | 40.1 | 30.0 | 20 | 79 | 748 | 4,157 | 41.8 | 80.0 | 31.2 | 29.6 | 35.6 | 65.5 | 32.3 | 61.7 | 73.9 | 1.7 | 200 | 283 |
| ETH-Crossing | 26.9 | 30.4 | 23.2 | 1 | 18 | 22 | 706 | 29.6 | 93.1 | 23.5 | 22.8 | 26.3 | 70.0 | 23.7 | 74.6 | 81.1 | 0.1 | 5 | 5 |
| ETH-Jelmoli | 51.1 | 57.6 | 42.1 | 14 | 14 | 339 | 869 | 65.7 | 83.1 | 41.0 | 43.3 | 46.7 | 68.7 | 50.9 | 64.3 | 79.4 | 0.8 | 33 | 41 |
| ETH-Linthescher | 35.4 | 40.4 | 31.3 | 17 | 116 | 143 | 5,572 | 37.6 | 95.9 | 34.2 | 28.8 | 37.0 | 75.7 | 29.8 | 76.0 | 80.1 | 0.1 | 55 | 64 |
| KITTI-16 | 41.3 | 42.0 | 31.4 | 2 | 1 | 311 | 646 | 62.0 | 77.2 | 25.1 | 39.7 | 30.7 | 46.2 | 47.5 | 59.1 | 76.3 | 1.5 | 42 | 57 |
| KITTI-19 | 30.8 | 44.5 | 30.8 | 7 | 14 | 1,025 | 2,579 | 51.7 | 72.9 | 29.8 | 32.3 | 34.4 | 58.0 | 38.1 | 53.6 | 72.5 | 1.0 | 95 | 160 |
| PETS09-S2L2 | 40.0 | 29.3 | 23.4 | 4 | 6 | 678 | 4,885 | 49.3 | 87.5 | 15.9 | 34.7 | 17.3 | 57.1 | 37.4 | 66.4 | 75.6 | 1.6 | 222 | 255 |
| TUD-Crossing | 75.5 | 63.8 | 45.5 | 8 | 0 | 43 | 209 | 81.0 | 95.4 | 36.0 | 57.6 | 43.9 | 58.5 | 62.6 | 73.7 | 79.3 | 0.2 | 18 | 18 |
| Venice-1 | -26.5 | 23.7 | 23.3 | 5 | 3 | 3,850 | 1,852 | 59.4 | 41.3 | 21.5 | 25.8 | 23.9 | 59.2 | 46.6 | 32.4 | 73.3 | 8.6 | 70 | 60 |
Raw data:
