MHT_bLSTM: MHT with Neural Gating using Bilinear LSTM
MOT17-01-SDP
Benchmark:
MOT17 |
Short name:
MHT_bLSTM
Detector:
Public
Description:
n/a
Reference:
C. Kim, F. Li, J. Rehg. Multi-object Tracking with Neural Gating Using Bilinear LSTM. In ECCV, 2018.
Last submitted:
July 26, 2018 (5 years ago)
Published:
July 26, 2018 at 08:47:57 CET
Submissions:
2
Project page / code:
n/a
Open source:
No
Hardware:
Titan X GPU, 3.6GHZ CPU
Runtime:
1.9 Hz
Benchmark performance:
| Sequence | MOTA | IDF1 | HOTA | MT | ML | FP | FN | Rcll | Prcn | AssA | DetA | AssRe | AssPr | DetRe | DetPr | LocA | FAF | ID Sw. | Frag |
| MOT17 | 47.5 | 51.9 | 41.0 | 429 (18.2) | 981 (41.7) | 25,981 | 268,042 | 52.5 | 91.9 | 42.8 | 39.5 | 47.3 | 73.8 | 42.2 | 73.8 | 80.4 | 1.5 | 2,069 (39.4) | 3,124 (59.5) |
Detailed performance:
| Sequence | MOTA | IDF1 | HOTA | MT | ML | FP | FN | Rcll | Prcn | AssA | DetA | AssRe | AssPr | DetRe | DetPr | LocA | FAF | ID Sw. | Frag |
| MOT17-01-DPM | 27.3 | 36.3 | 28.0 | 5 | 10 | 785 | 3,868 | 40.0 | 76.7 | 29.0 | 27.3 | 35.0 | 61.6 | 30.6 | 58.7 | 75.7 | 1.7 | 39 | 63 |
| MOT17-01-FRCNN | 31.9 | 50.7 | 41.7 | 7 | 10 | 1,279 | 3,099 | 52.0 | 72.4 | 49.1 | 35.7 | 55.3 | 74.2 | 42.8 | 59.6 | 78.7 | 2.8 | 14 | 37 |
| MOT17-01-SDP | 43.5 | 53.1 | 41.7 | 8 | 7 | 455 | 3,154 | 51.1 | 87.9 | 47.0 | 37.2 | 53.9 | 72.9 | 40.4 | 69.5 | 78.9 | 1.0 | 34 | 56 |
| MOT17-03-DPM | 49.4 | 53.3 | 40.3 | 32 | 30 | 7,279 | 45,363 | 56.7 | 89.1 | 39.7 | 41.0 | 42.3 | 74.5 | 44.7 | 70.3 | 79.1 | 4.9 | 301 | 526 |
| MOT17-03-FRCNN | 57.0 | 58.1 | 44.9 | 38 | 28 | 2,949 | 41,907 | 60.0 | 95.5 | 44.5 | 45.7 | 47.4 | 76.6 | 48.2 | 76.8 | 81.0 | 2.0 | 188 | 259 |
| MOT17-03-SDP | 72.5 | 67.1 | 53.4 | 69 | 15 | 1,981 | 26,541 | 74.6 | 97.5 | 49.9 | 57.4 | 52.8 | 77.1 | 60.3 | 78.8 | 81.4 | 1.3 | 228 | 426 |
| MOT17-06-DPM | 41.7 | 44.9 | 34.9 | 39 | 108 | 1,048 | 5,734 | 51.3 | 85.2 | 33.8 | 36.2 | 56.4 | 50.7 | 40.1 | 66.5 | 78.1 | 0.9 | 93 | 131 |
| MOT17-06-FRCNN | 47.6 | 50.0 | 40.4 | 51 | 73 | 1,036 | 5,038 | 57.2 | 86.7 | 37.9 | 43.1 | 58.0 | 56.7 | 47.6 | 72.1 | 81.8 | 0.9 | 101 | 139 |
| MOT17-06-SDP | 51.0 | 54.5 | 42.6 | 66 | 81 | 942 | 4,727 | 59.9 | 88.2 | 41.7 | 43.7 | 58.4 | 60.3 | 48.3 | 71.1 | 81.0 | 0.8 | 110 | 118 |
| MOT17-07-DPM | 33.8 | 40.1 | 31.5 | 4 | 31 | 483 | 10,579 | 37.4 | 92.9 | 35.4 | 28.1 | 37.4 | 73.7 | 29.3 | 72.8 | 78.5 | 1.0 | 114 | 160 |
| MOT17-07-FRCNN | 32.7 | 39.3 | 31.9 | 3 | 22 | 951 | 10,325 | 38.9 | 87.4 | 35.1 | 29.6 | 37.9 | 70.9 | 31.5 | 70.7 | 78.6 | 1.9 | 98 | 171 |
| MOT17-07-SDP | 40.5 | 43.4 | 35.2 | 6 | 24 | 722 | 9,224 | 45.4 | 91.4 | 36.0 | 34.5 | 39.1 | 69.6 | 36.7 | 73.8 | 80.4 | 1.4 | 109 | 153 |
| MOT17-08-DPM | 26.7 | 31.7 | 27.9 | 11 | 38 | 613 | 14,775 | 30.1 | 91.2 | 32.8 | 23.8 | 36.1 | 71.4 | 24.8 | 75.2 | 81.6 | 1.0 | 86 | 110 |
| MOT17-08-FRCNN | 21.8 | 31.1 | 27.7 | 8 | 43 | 710 | 15,755 | 25.4 | 88.3 | 37.9 | 20.4 | 43.0 | 74.6 | 21.3 | 73.9 | 82.4 | 1.1 | 46 | 58 |
| MOT17-08-SDP | 28.9 | 36.0 | 32.4 | 12 | 37 | 718 | 14,189 | 32.8 | 90.6 | 40.6 | 25.9 | 44.9 | 73.1 | 27.2 | 75.1 | 82.1 | 1.1 | 102 | 124 |
| MOT17-12-DPM | 38.9 | 48.9 | 38.3 | 13 | 44 | 563 | 4,701 | 45.8 | 87.6 | 43.2 | 34.1 | 54.6 | 67.0 | 36.9 | 70.5 | 80.1 | 0.6 | 31 | 43 |
| MOT17-12-FRCNN | 34.4 | 48.1 | 37.2 | 11 | 49 | 542 | 5,129 | 40.8 | 86.7 | 44.8 | 30.9 | 55.0 | 67.6 | 33.5 | 71.1 | 81.8 | 0.6 | 17 | 23 |
| MOT17-12-SDP | 43.4 | 53.1 | 43.2 | 19 | 47 | 417 | 4,466 | 48.5 | 91.0 | 50.0 | 37.5 | 60.0 | 71.2 | 40.1 | 75.2 | 82.5 | 0.5 | 22 | 28 |
| MOT17-14-DPM | 21.7 | 31.0 | 23.4 | 8 | 103 | 609 | 13,779 | 25.5 | 88.5 | 28.8 | 19.1 | 30.3 | 75.2 | 19.9 | 69.3 | 78.4 | 0.8 | 92 | 117 |
| MOT17-14-FRCNN | 23.5 | 35.4 | 26.6 | 12 | 87 | 1,453 | 12,557 | 32.1 | 80.3 | 30.8 | 23.4 | 34.3 | 68.3 | 25.3 | 63.3 | 76.1 | 1.9 | 136 | 248 |
| MOT17-14-SDP | 26.0 | 34.1 | 25.2 | 7 | 94 | 446 | 13,132 | 29.0 | 92.3 | 29.4 | 21.8 | 31.8 | 74.9 | 22.7 | 72.3 | 79.5 | 0.6 | 108 | 134 |
Raw data:
