Graph Neural Network for 2D and 3D Multi-Object Tracking


Short name:

GNNMOT

Benchmark:

Description:

n/a

Hardware:

2.5HZ, 6 Cores

Detector:

Public

Processing:

Online

Last submitted:

October 31, 2019 (13 days ago)

Published:

October 31, 2019 at 15:05:09 CET

Submissions:

1

Open source:

Yes

Project page / code:

Reference:

Anonymous submission

Benchmark performance:

MOTAMOTPFAFMTMLFPFNID Sw.FragSpecificationsDetector
42.077.01.312.0 % 50.0 % 23,294299,6944,3773,8472.5HZ, 6 CoresPublic
IDF1ID PrecisionID Recall
29.343.422.1

Detailed performance:

Sequence MOTA IDF1 MOTP FAF GT MT ML FP FN ID Sw Frag
MOT17-01-DPM22.020.472.60.2248.3 % 58.3 % 1114,8892938
MOT17-03-DPM37.018.675.44.914810.8 % 29.1 % 7,27957,703938616
MOT17-06-DPM31.032.374.30.32226.8 % 65.3 % 3687,65810479
MOT17-07-DPM26.220.573.81.2603.3 % 61.7 % 60411,727132124
MOT17-08-DPM20.119.578.51.4766.6 % 56.6 % 87715,87412587
MOT17-12-DPM32.533.776.50.6918.8 % 60.4 % 5035,2776648
MOT17-14-DPM11.214.274.40.21640.6 % 84.1 % 13116,2165969
MOT17-01-FRCNN32.336.477.22.02420.8 % 45.8 % 8803,4583145
MOT17-03-FRCNN53.134.878.41.514819.6 % 20.3 % 2,21146,180653451
MOT17-06-FRCNN41.743.180.40.222213.1 % 52.3 % 2536,5299285
MOT17-07-FRCNN30.931.376.40.8603.3 % 46.7 % 37711,187112180
MOT17-08-FRCNN21.427.281.50.5766.6 % 60.5 % 32316,2225051
MOT17-12-FRCNN34.442.579.60.29112.1 % 61.5 % 1485,5082719
MOT17-14-FRCNN17.222.673.00.91640.6 % 67.7 % 70114,466132167
MOT17-01-SDP41.733.773.61.82433.3 % 20.8 % 8232,8548499
MOT17-03-SDP69.334.577.32.914844.6 % 8.8 % 4,36326,7991,007924
MOT17-06-SDP45.543.777.00.522219.4 % 40.1 % 5875,668169142
MOT17-07-SDP39.428.775.42.06010.0 % 33.3 % 9899,053191203
MOT17-08-SDP28.426.078.40.87613.2 % 46.1 % 47314,521129157
MOT17-12-SDP40.242.478.80.49117.6 % 50.5 % 3614,7665839
MOT17-14-SDP22.825.573.21.21641.8 % 59.1 % 93213,139189224

Raw data:

n/a


GNNMOT
dps