Benchmark:
Short name:
GMPHD_SAF
Description:
n/a
Reference:
Y. Song, M. Jeon. Online Multi-Object Tracking and Segmentation with GMPHD Filter and Simple Affinity Fusion. In arXiv preprint arXiv:2009.00100, 2020.
Last submitted:
June 01, 2020 (3 years ago)
Published:
May 30, 2020 at 00:43:44 CET
Submissions:
2
Project page / code:
n/a
Open source:
No
Hardware:
Intel i7-7700K @ 4.2GHz, 4 Cores
Runtime:
3.8 Hz
Benchmark performance:
| Sequence | sMOTSA | IDF1 | MOTSA | MOTSP | MODSA | MT | ML | TP | FP | FN | Rcll | Prcn | ID Sw. | Frag |
| CVPR 2020 MOTS Challenge | 68.4 | 64.9 | 82.6 | 83.9 | 84.4 | 248 (75.6) | 10 (3.0) | 28,382 | 1,161 | 3,887 | 88.0 | 96.1 | 569 (646.9) | 770 (875.5) |
Detailed performance:
| Sequence | sMOTSA | IDF1 | MOTSA | MOTSP | MODSA | MT | ML | TP | FP | FN | Rcll | Prcn | ID Sw. | Frag |
| MOTS20-01 | 69.0 | 70.6 | 84.4 | 82.3 | 85.0 | 9 | 0 | 2,726 | 87 | 380 | 87.8 | 96.9 | 16 | 58 |
| MOTS20-06 | 72.9 | 63.8 | 86.8 | 85.1 | 90.5 | 155 | 4 | 9,127 | 250 | 687 | 93.0 | 97.3 | 363 | 384 |
| MOTS20-07 | 64.5 | 59.7 | 80.1 | 81.5 | 81.0 | 36 | 4 | 10,830 | 393 | 2,048 | 84.1 | 96.5 | 127 | 231 |
| MOTS20-12 | 69.1 | 73.7 | 80.4 | 87.1 | 81.4 | 48 | 2 | 5,699 | 431 | 772 | 88.1 | 93.0 | 63 | 97 |
Raw data: