MDP Tracking with SubCNN detections

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

Short name:

MDP_SubCNN

Benchmark:

Description:

We used the SubCNN detections with the MDP tracker.
Tracking:
Yu Xiang, Alexandre Alahi and Silvio Savarese. Learning to Track: Online Multi-Object Tracking by Decision Making. In International Conference on Computer Vision (ICCV), pp. 4705-4713, 2015.
Detection:
Yu Xiang, Wongun Choi, Yuanqing Lin and Silvio Savarese. Subcategory-aware Convolutional Neural Networks for Object Proposals and Detection. arXiv:1604.04693, 2016.

Hardware:

3.5 Ghz, 8 cores

Detector:

Private

Processing:

Online

Last submitted:

June 28, 2016 (3 years ago)

Published:

June 24, 2016 at 08:14:41 CET

Submissions:

2

Open source:

Yes

Project page / code:

Reference:

Y. Xiang, A. Alahi, S. Savarese. Learning to Track: Online Multi-Object Tracking by Decision Making. In International Conference on Computer Vision (ICCV), 2015.

Benchmark performance:

MOTAMOTPFAFMTMLFPFNID Sw.FragSpecificationsDetector
47.574.21.530.0 % 18.6 % 8,63122,9696281,3703.5 Ghz, 8 coresPrivate
IDF1ID PrecisionID Recall
55.764.249.2

Detailed performance:

Sequence MOTA IDF1 MOTP FAF GT MT ML FP FN ID Sw Frag
TUD-Crossing78.974.576.70.21369.2 % 0.0 % 32195616
PETS09-S2L247.536.872.60.8427.1 % 9.5 % 3414,524196332
ETH-Jelmoli48.265.777.31.14535.6 % 22.2 % 492814937
ETH-Linthescher63.967.177.10.419724.4 % 31.0 % 4952,65770143
ETH-Crossing63.876.579.50.32619.2 % 26.9 % 64293621
AVG-TownCentre49.564.570.13.122638.9 % 15.5 % 1,3812,106121297
ADL-Rundle-133.449.972.45.83234.4 % 0.0 % 2,8993,23070142
ADL-Rundle-344.951.679.61.34420.5 % 15.9 % 7934,7525699
KITTI-1650.066.670.31.31735.3 % 5.9 % 2625662247
KITTI-1940.961.868.11.16224.2 % 8.1 % 1,1431,96551187
Venice-142.648.476.01.61735.3 % 23.5 % 7291,8672149

Raw data:


POI
MDP_SubCNN