HAM_SADF: Historical Appearance Matching and Scene Adaptive Detection Filtering
Video not available.
Rendering of new sequences is currently deactivated due to heavy load.
Rendering of new sequences is currently deactivated due to heavy load.
Rendering of new sequences is currently deactivated due to heavy load.
Rendering of new sequences is currently deactivated due to heavy load.
Rendering of new sequences is currently deactivated due to heavy load.
Rendering of new sequences is currently deactivated due to heavy load.
Rendering of new sequences is currently deactivated due to heavy load.
Rendering of new sequences is currently deactivated due to heavy load.
Rendering of new sequences is currently deactivated due to heavy load.
Rendering of new sequences is currently deactivated due to heavy load.
Rendering of new sequences is currently deactivated due to heavy load.
Benchmark:
MOT15 |
Short name:
HAM_SADF
Detector:
Public
Description:
https://github.com/yyc9268/Deep-TAMA
Reference:
Y. Yoon, A. Boragule, Y. Song, K. Yoon, M. Jeon. Online Multi-Object Tracking with Historical Appearance Matching and Scene Adaptive Detection Filtering. In IEEE AVSS, 2018.
Last submitted:
June 13, 2018 (6 years ago)
Published:
June 04, 2018 at 07:12:57 CET
Submissions:
3
Project page / code:
Open source:
Yes
Hardware:
3.7 GHZ, 4 Core, TITAN X
Runtime:
18.7 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 | 25.2 | 37.8 | 28.2 | 41 (5.7) | 420 (58.3) | 7,330 | 38,275 | 37.7 | 76.0 | 31.5 | 25.5 | 34.2 | 68.5 | 28.7 | 57.8 | 75.5 | 1.3 | 357 (9.5) | 745 (19.8) |
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 | 11.9 | 38.0 | 29.0 | 6 | 8 | 3,449 | 4,695 | 49.5 | 57.2 | 30.9 | 27.8 | 34.5 | 65.2 | 38.1 | 43.9 | 75.3 | 6.9 | 56 | 121 |
| ADL-Rundle-3 | 33.2 | 42.0 | 31.1 | 4 | 11 | 1,259 | 5,468 | 46.2 | 78.9 | 31.4 | 31.0 | 33.2 | 71.0 | 35.2 | 60.0 | 76.6 | 2.0 | 60 | 61 |
| AVG-TownCentre | 14.5 | 26.5 | 20.6 | 5 | 158 | 242 | 5,841 | 18.3 | 84.4 | 31.8 | 13.5 | 34.8 | 66.7 | 13.9 | 64.4 | 74.4 | 0.5 | 27 | 78 |
| ETH-Crossing | 26.1 | 41.4 | 27.3 | 1 | 18 | 22 | 718 | 28.4 | 92.8 | 35.0 | 21.3 | 38.0 | 74.9 | 22.1 | 72.2 | 79.0 | 0.1 | 1 | 2 |
| ETH-Jelmoli | 37.9 | 57.0 | 37.9 | 7 | 21 | 161 | 1,408 | 44.5 | 87.5 | 44.8 | 32.2 | 50.0 | 71.0 | 34.6 | 68.1 | 77.8 | 0.4 | 7 | 30 |
| ETH-Linthescher | 15.5 | 23.6 | 21.0 | 6 | 161 | 84 | 7,448 | 16.6 | 94.6 | 35.5 | 12.4 | 38.1 | 72.6 | 12.7 | 72.1 | 77.5 | 0.1 | 11 | 35 |
| KITTI-16 | 38.3 | 47.8 | 29.8 | 0 | 4 | 88 | 944 | 44.5 | 89.6 | 28.1 | 31.7 | 31.8 | 56.0 | 33.9 | 68.2 | 76.9 | 0.4 | 18 | 33 |
| KITTI-19 | 26.0 | 43.8 | 30.4 | 4 | 21 | 746 | 3,160 | 40.9 | 74.5 | 34.0 | 27.7 | 37.7 | 65.1 | 30.9 | 56.3 | 72.1 | 0.7 | 48 | 142 |
| PETS09-S2L2 | 39.2 | 37.6 | 27.0 | 3 | 10 | 549 | 5,216 | 45.9 | 89.0 | 22.6 | 32.5 | 24.0 | 65.8 | 34.3 | 66.5 | 74.2 | 1.3 | 101 | 189 |
| TUD-Crossing | 58.9 | 68.1 | 47.9 | 4 | 2 | 72 | 372 | 66.2 | 91.0 | 50.1 | 45.9 | 52.5 | 73.5 | 50.4 | 69.3 | 78.4 | 0.4 | 9 | 13 |
| Venice-1 | 19.3 | 33.1 | 25.8 | 1 | 6 | 658 | 3,005 | 34.1 | 70.3 | 29.9 | 22.5 | 31.2 | 74.0 | 26.0 | 53.5 | 76.4 | 1.5 | 19 | 41 |
Raw data:
