Recurrent Autoregressive Networks for Online Multi-Object Tracking

TUD-Crossing PETS09-S2L2 ETH-Jelmoli ETH-Linthescher ETH-Crossing AVG-TownCentre ADL-Rundle-1 ADL-Rundle-3 KITTI-16 KITTI-19

Short name:

RAR15pub

Benchmark:

Description:

Recurrent autoregressive networks tracking framework with public detections

Hardware:

Titan X, 1.5GHZ, 1Core

Detector:

Public

Processing:

Online

Last submitted:

March 27, 2017 (8 months ago)

Published:

April 22, 2017 at 09:35:33 CET

Submissions:

2

Open source:

No

Project page / code:

n/a

Reference:

Anonymous ICCV submission

Benchmark performance:

MOTAMOTPFAFMTMLFPFNID Sw.FragSpecificationsDetector
35.170.91.213.0 % 42.3 % 6,77132,7173811,523Titan X, 1.5GHZ, 1CorePublic
IDF1ID PrecisionID Recall
45.462.035.8

Detailed performance:

Sequence MOTA IDF1 MOTP FAF GT MT ML FP FN ID Sw Frag
TUD-Crossing70.177.872.70.21346.2 % 7.7 % 462731129
PETS09-S2L249.636.469.61.8429.5 % 7.1 % 7983,923135357
ETH-Jelmoli40.658.372.71.14520.0 % 28.9 % 4861,0041879
ETH-Linthescher26.938.273.70.21976.1 % 67.0 % 1996,3072364
ETH-Crossing37.655.174.80.02619.2 % 53.8 % 761900
AVG-TownCentre30.351.369.42.022618.6 % 42.5 % 9174,03825237
ADL-Rundle-126.347.471.54.33225.0 % 34.4 % 2,1354,69031168
ADL-Rundle-339.643.172.21.2449.1 % 22.7 % 7295,34464148
KITTI-1641.257.772.50.8170.0 % 17.6 % 1748091866
KITTI-1931.350.166.10.8624.8 % 22.6 % 8192,80941300
Venice-126.039.071.81.0175.9 % 47.1 % 4612,9011575

Raw data:

n/a


FWT
RAR15pub