PHDTT: Pedestrian Heads Detection and Tracking with Transformer
HT21-11
Benchmark:
Short name:
PHDTT
Description:
Reference:
Vo, Xuan-Thuy, et al. "Pedestrian Head Detection and Tracking via Global Vision Transformer." International Workshop on Frontiers of Computer Vision. Springer, Cham, 2022.
Last submitted:
September 06, 2021 (2 years ago)
Published:
August 07, 2021 at 09:52:06 CET
Submissions:
3
Project page / code:
n/a
Open source:
No
Hardware:
Tesla V100
Runtime:
21.2 Hz
Benchmark performance:
| Sequence | MOTA | IDF1 | IDEucl | HOTA | MT | ML | FP | FN | Rcll | Prcn | DetPr | DetRe | ID Sw. | Frag |
| Head Tracking 21 | 60.6 | 47.9 | 52.6 | 36.1 | 1,104 (47.1) | 188 (8.0) | 132,714 | 184,215 | 78.1 | 83.2 | 57.7 | 54.2 | 15,004 (0.0) | 41,556 (0.0) |
Detailed performance:
| Sequence | MOTA | IDF1 | IDEucl | HOTA | MT | ML | FP | FN | Rcll | Prcn | DetPr | DetRe | ID Sw. | Frag |
| HT21-11 | 79.0 | 63.7 | 65.6 | 47.8 | 84 | 4 | 1,485 | 4,620 | 84.7 | 94.5 | 69.7 | 62.4 | 232 | 1,254 |
| HT21-12 | 72.0 | 57.6 | 63.1 | 40.7 | 514 | 11 | 36,824 | 48,584 | 84.5 | 87.8 | 58.9 | 56.7 | 2,352 | 13,883 |
| HT21-13 | 37.3 | 24.9 | 27.0 | 20.2 | 192 | 63 | 66,256 | 60,498 | 72.3 | 70.4 | 47.9 | 49.2 | 9,962 | 19,005 |
| HT21-14 | 68.4 | 59.7 | 55.1 | 43.4 | 223 | 59 | 11,514 | 46,604 | 75.0 | 92.4 | 68.1 | 55.2 | 838 | 2,551 |
| HT21-15 | 55.1 | 44.3 | 52.4 | 33.1 | 91 | 51 | 16,635 | 23,909 | 74.5 | 80.8 | 56.8 | 52.4 | 1,620 | 4,863 |
Raw data:
