GNNMatch: Graph Convolutional Neural Networks for Multi-Object Tracking via Sinkhorn Normalization
MOT16-01
Short name:
GNNMatch
Detector:
Public
Description:
Reference:
I. Papakis, A. Sarkar, A. Karpatne. GCNNMatch: Graph Convolutional Neural Networks for Multi-Object Tracking via Sinkhorn Normalization. In , 2020.
Last submitted:
July 26, 2020 (3 months ago)
Published:
July 30, 2020 at 21:33:18 CET
Submissions:
1
Project page / code:
Open source:
Yes
Hardware:
TITAN RTX
Runtime:
0.3 Hz
Benchmark performance:
| Sequence | MOTA | IDF1 | MOTP | MT | ML | FP | FN | Recall | Precision | FAF | ID Sw. | Frag |
| MOT16 | 56.9 | 55.9 | 79.1 | 169 (22.3) | 268 (35.3) | 3,235 | 74,784 | 59.0 | 97.1 | 0.5 | 564 (9.6) | 818 (13.9) |
Detailed performance:
| Sequence | MOTA | IDF1 | MOTP | MT | ML | FP | FN | Recall | Precision | FAF | ID Sw. | Frag |
| MOT16-01 | 43.6 | 36.1 | 77.9 | 6 | 10 | 58 | 3,526 | 44.9 | 98.0 | 0.1 | 20 | 27 |
| MOT16-03 | 68.3 | 65.0 | 78.9 | 61 | 19 | 1,833 | 31,200 | 70.2 | 97.6 | 1.2 | 122 | 225 |
| MOT16-06 | 54.3 | 38.5 | 80.1 | 59 | 84 | 502 | 4,666 | 59.6 | 93.2 | 0.4 | 108 | 133 |
| MOT16-07 | 45.5 | 47.2 | 78.4 | 7 | 14 | 221 | 8,591 | 47.4 | 97.2 | 0.4 | 79 | 133 |
| MOT16-08 | 34.5 | 34.4 | 82.4 | 8 | 25 | 205 | 10,702 | 36.1 | 96.7 | 0.3 | 58 | 77 |
| MOT16-12 | 48.3 | 55.0 | 82.1 | 17 | 39 | 109 | 4,159 | 49.9 | 97.4 | 0.1 | 20 | 30 |
| MOT16-14 | 32.9 | 39.7 | 76.9 | 11 | 77 | 307 | 11,940 | 35.4 | 95.5 | 0.4 | 157 | 193 |
Raw data:
