BiGRU1: Framework of Biconnected GRU with Fully-connected Networks

TUD-Crossing


Benchmark:

Short name:

BiGRU1

Detector:

Public

Description:

n/a

Reference:

Zaifeng Shi, Huizheng Ren, Qingjie Cao, Boyu Fan, and Qiangqiang FanData association framework based on biconnected gated recurrent unit network for multiple object tracking. In JEI, 2020.

Last submitted:

November 11, 2019 (1 year ago)

Published:

November 28, 2020 at 10:05:00 CET

Submissions:

1

Project page / code:

n/a

Open source:

No

Hardware:

E5 1650, 24G DDR3, and gtx 1070

Runtime:

4.0 Hz

Benchmark performance:

Sequence MOTA IDF1 MOTP MT ML FP FN Recall Precision FAF ID Sw. Frag
2D MOT 201526.132.271.247 (6.5)352 (48.8)5,76138,94836.679.61.0719 (19.6)2,046 (55.9)

Detailed performance:

Sequence MOTA IDF1 MOTP MT ML FP FN Recall Precision FAF ID Sw. Frag
ADL-Rundle-18.231.570.52112,8765,57740.156.55.894381
ADL-Rundle-338.242.772.94141776,07040.395.90.338207
AVG-TownCentre19.88.969.781016044,97930.378.21.3147349
ETH-Crossing27.232.874.81172170529.793.40.148
ETH-Jelmoli42.350.573.311133151,11955.981.80.72965
ETH-Linthescher16.422.975.58155897,36617.594.60.1959
KITTI-1635.936.371.3036999941.391.10.32270
KITTI-1929.940.167.63185603,11541.779.90.573271
PETS09-S2L232.526.369.41106485,56842.286.31.5291575
TUD-Crossing73.767.473.0928519782.191.40.4810
Venice-121.739.772.9083173,25328.780.50.7451

Raw data: