UN_DAM: Multi Object Tracking using Deep Structural Cost Minimization in Data Association

TUD-Crossing


Benchmark:

Short name:

UN_DAM

Detector:

Public

Description:

n/a

Reference:

Multi Object Tracking using Deep Structural Cost Minimization in Data Association

Last submitted:

June 09, 2019 (1 year ago)

Published:

June 14, 2019 at 13:41:22 CET

Submissions:

2

Project page / code:

n/a

Open source:

No

Hardware:

3 GHZ, 1 Core GTx 1080

Runtime:

20.7 Hz

Benchmark performance:

Sequence MOTA IDF1 MOTP MT ML FP FN Recall Precision FAF ID Sw. Frag
2D MOT 201529.741.471.466 (9.2)360 (49.9)7,61035,26942.677.51.3318 (7.5)674 (15.8)

Detailed performance:

Sequence MOTA IDF1 MOTP MT ML FP FN Recall Precision FAF ID Sw. Frag
ADL-Rundle-119.342.772.59102,9424,52751.461.95.94165
ADL-Rundle-335.039.774.13159415,62044.782.91.54946
AVG-TownCentre23.144.870.1221168914,57736.074.32.027134
ETH-Crossing29.344.174.22161569231.095.40.123
ETH-Jelmoli39.056.972.48163061,23451.481.00.7839
ETH-Linthescher18.927.874.071491177,11420.393.90.11421
KITTI-1633.248.173.4051201,00540.985.30.61115
KITTI-1924.745.964.82188973,09442.171.50.931155
PETS09-S2L248.340.269.2666214,25655.989.71.4111167
TUD-Crossing60.368.675.1536636766.791.80.357
Venice-123.435.872.1266942,78339.071.91.51922

Raw data: