Benchmark:
MOT15 |
Short name:
AP_HWDPL_p
Detector:
Public
Description:
Online tracking with appearance model based on R-CNN from Huawei (HWDPL).
Reference:
C. Long, A. Haizhou, S. Chong, Z. Zijie, B. Bo. Online Multi-Object Tracking with Convolutional Neural Networks. In 2017 IEEE International Conference on Image Processing (ICIP), 2017.
Last submitted:
December 13, 2016 (7 years ago)
Published:
September 26, 2017 at 09:20:18 CET
Submissions:
2
Project page / code:
n/a
Open source:
No
Hardware:
GTX 1080, 2.3 GHz, 1 Core
Runtime:
6.7 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 | 38.5 | 47.1 | 35.0 | 63 (8.7) | 270 (37.4) | 4,005 | 33,203 | 46.0 | 87.6 | 37.9 | 32.9 | 42.4 | 65.0 | 35.4 | 67.4 | 76.7 | 0.7 | 586 (12.8) | 1,263 (27.5) |
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 | 35.3 | 43.6 | 34.6 | 4 | 10 | 573 | 5,413 | 41.8 | 87.2 | 41.3 | 29.4 | 44.1 | 71.9 | 31.7 | 66.0 | 76.7 | 1.1 | 39 | 81 |
| ADL-Rundle-3 | 43.3 | 51.9 | 39.8 | 7 | 9 | 643 | 5,076 | 50.1 | 88.8 | 43.0 | 37.2 | 47.1 | 70.7 | 40.5 | 71.7 | 81.0 | 1.0 | 47 | 85 |
| AVG-TownCentre | 28.4 | 44.7 | 31.6 | 9 | 63 | 941 | 4,005 | 44.0 | 77.0 | 32.0 | 31.4 | 36.9 | 56.5 | 34.1 | 59.7 | 71.7 | 2.1 | 169 | 412 |
| ETH-Crossing | 34.2 | 42.3 | 30.6 | 1 | 16 | 20 | 630 | 37.2 | 94.9 | 31.3 | 30.0 | 32.7 | 72.9 | 30.9 | 78.8 | 81.7 | 0.1 | 10 | 13 |
| ETH-Jelmoli | 52.9 | 65.5 | 45.2 | 10 | 13 | 178 | 1,001 | 60.5 | 89.6 | 47.6 | 42.9 | 54.9 | 66.6 | 47.2 | 69.9 | 79.6 | 0.4 | 16 | 37 |
| ETH-Linthescher | 39.4 | 47.5 | 36.5 | 17 | 124 | 185 | 5,185 | 41.9 | 95.3 | 43.2 | 31.1 | 50.3 | 62.6 | 32.4 | 73.6 | 78.8 | 0.2 | 44 | 71 |
| KITTI-16 | 40.7 | 61.1 | 37.3 | 3 | 2 | 160 | 831 | 51.1 | 84.5 | 41.2 | 34.0 | 43.8 | 68.1 | 37.3 | 61.7 | 74.1 | 0.8 | 18 | 35 |
| KITTI-19 | 34.3 | 50.6 | 33.5 | 2 | 20 | 663 | 2,805 | 47.5 | 79.3 | 34.7 | 32.8 | 38.4 | 61.0 | 36.0 | 60.1 | 72.6 | 0.6 | 43 | 147 |
| PETS09-S2L2 | 38.9 | 34.3 | 24.1 | 1 | 4 | 552 | 5,164 | 46.4 | 89.0 | 17.9 | 32.8 | 20.5 | 49.5 | 34.9 | 67.0 | 75.0 | 1.3 | 179 | 328 |
| TUD-Crossing | 61.3 | 64.1 | 44.8 | 5 | 2 | 14 | 401 | 63.6 | 98.0 | 44.7 | 45.0 | 50.9 | 68.2 | 47.5 | 73.2 | 77.9 | 0.1 | 12 | 27 |
| Venice-1 | 39.1 | 48.7 | 35.9 | 4 | 7 | 76 | 2,692 | 41.0 | 96.1 | 44.1 | 29.3 | 47.2 | 69.9 | 30.3 | 71.1 | 76.4 | 0.2 | 9 | 27 |
Raw data: