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:

RAR15

Benchmark:

Description:

Recurrent autoregressive networks tracking framework with faster R-CNN detections

Hardware:

Titan X, 1.5GHZ, 1Core

Detector:

Private

Processing:

Online

Last submitted:

March 13, 2017 (2 years ago)

Published:

July 23, 2018 at 09:30:04 CET

Submissions:

1

Open source:

No

Project page / code:

Reference:

K. Fang, Y. Xiang, X. Li, S. Savarese. Recurrent Autoregressive Networks for Online Multi-Object Tracking. In The IEEE Winter Conference on Applications of Computer Vision (WACV), 2018.

Benchmark performance:

MOTAMOTPFAFMTMLFPFNID Sw.FragSpecificationsDetector
56.573.01.645.1 % 14.6 % 9,38616,9214281,364Titan X, 1.5GHZ, 1CorePrivate
IDF1ID PrecisionID Recall
61.365.657.6

Detailed performance:

Sequence MOTA IDF1 MOTP FAF GT MT ML FP FN ID Sw Frag
TUD-Crossing91.686.978.30.11384.6 % 0.0 % 167345
PETS09-S2L267.843.073.11.74231.0 % 2.4 % 7422,205159291
ETH-Jelmoli58.774.975.81.04548.9 % 22.2 % 441599737
ETH-Linthescher70.576.475.90.819746.7 % 25.4 % 9101,68932100
ETH-Crossing73.184.479.60.72653.8 % 15.4 % 150117313
AVG-TownCentre66.277.569.61.922648.7 % 11.9 % 8421,54231265
ADL-Rundle-132.650.270.57.13253.1 % 0.0 % 3,5532,65963189
ADL-Rundle-349.848.977.11.24427.3 % 11.4 % 7344,3066787
KITTI-1642.865.867.41.81735.3 % 5.9 % 3695871769
KITTI-1942.068.267.21.36235.5 % 6.5 % 1,3611,70633255
Venice-162.364.173.00.61735.3 % 17.6 % 2681,4381253

Raw data:


AM
RAR15