Short name:
BrinqTraq_v2
Detector:
Private
Description:
Custom trained yolo_v5 x-large model with input size of 1920x1920 as detector.
Reference:
H. Mahadik, N. Bamra, J. Rynne. Brinq Edge AI SDK v2.0, Arcturus Networks Inc. In , 2023.
Last submitted:
April 14, 2023 (1 year ago)
Published:
April 14, 2023 at 19:08:32 CET
Submissions:
3
Project page / code:
Open source:
No
Hardware:
1.9 GHz, 2560 CUDA Cores
Runtime:
6.1 Hz
Benchmark performance:
| Sequence | MOTA | IDF1 | HOTA | MT | ML | FP | FN | Rcll | Prcn | AssA | DetA | AssRe | AssPr | DetRe | DetPr | LocA | FAF | ID Sw. | Frag |
| MOT16 | 81.3 | 80.0 | 63.9 | 453 (59.7) | 62 (8.2) | 12,324 | 21,269 | 88.3 | 92.9 | 63.1 | 65.1 | 69.6 | 75.5 | 72.2 | 75.9 | 82.3 | 2.1 | 519 (0.0) | 1,185 (0.0) |
Detailed performance:
| Sequence | MOTA | IDF1 | HOTA | MT | ML | FP | FN | Rcll | Prcn | AssA | DetA | AssRe | AssPr | DetRe | DetPr | LocA | FAF | ID Sw. | Frag |
| MOT16-01 | 76.7 | 69.8 | 55.7 | 16 | 1 | 208 | 1,264 | 80.2 | 96.1 | 52.1 | 60.1 | 56.9 | 77.4 | 64.5 | 77.2 | 81.3 | 0.5 | 20 | 62 |
| MOT16-03 | 93.0 | 90.6 | 72.1 | 137 | 1 | 3,827 | 3,473 | 96.7 | 96.4 | 71.3 | 73.3 | 76.6 | 79.7 | 79.2 | 78.9 | 82.6 | 2.6 | 54 | 136 |
| MOT16-06 | 64.7 | 67.0 | 51.7 | 128 | 24 | 1,793 | 2,191 | 81.0 | 83.9 | 49.1 | 54.9 | 69.0 | 58.7 | 65.4 | 67.7 | 81.0 | 1.5 | 89 | 218 |
| MOT16-07 | 73.8 | 64.0 | 51.6 | 30 | 2 | 853 | 3,341 | 79.5 | 93.8 | 44.4 | 60.2 | 49.2 | 70.6 | 65.7 | 77.5 | 82.9 | 1.7 | 76 | 140 |
| MOT16-08 | 57.1 | 50.0 | 44.8 | 35 | 2 | 3,586 | 3,437 | 79.5 | 78.8 | 39.5 | 52.5 | 48.1 | 58.0 | 65.4 | 64.8 | 81.6 | 5.7 | 162 | 250 |
| MOT16-12 | 69.4 | 76.4 | 63.0 | 43 | 17 | 344 | 2,175 | 73.8 | 94.7 | 67.3 | 59.2 | 74.6 | 79.4 | 63.6 | 81.6 | 86.0 | 0.4 | 17 | 47 |
| MOT16-14 | 61.0 | 71.7 | 50.8 | 64 | 15 | 1,713 | 5,388 | 70.8 | 88.4 | 52.9 | 49.1 | 58.2 | 72.1 | 55.4 | 69.1 | 79.6 | 2.3 | 101 | 332 |
Raw data: