MotiCon: Learning an image-based motion context for multi-target tracking

TUD-Crossing


Benchmark:

Short name:

MotiCon

Detector:

Public

Description:

We present a novel method for multiple people tracking that leverages a generalized model for capturing interactions among individuals.
At the core of our model lies a learned dictionary of interaction feature strings which capture relationships between the motions of targets.
These feature strings, created from low-level image features, lead to a much richer representation of the physical interactions between targets compared to hand-specified social force models that previous works have introduced for tracking. One disadvantage of using social forces is that all pedestrians must be detected in order for the forces to be applied, while our method is able to encode the effect of undetected targets, making the tracker more robust to partial occlusions.
The interaction feature strings are used in a Random Forest framework to track targets according to the features surrounding them.

Reference:

L. Leal-Taixé, M. Fenzi, A. Kuznetsova, B. Rosenhahn, S. Savarese. Learning an image-based motion context for multiple people tracking. In CVPR, 2014.

Last submitted:

April 07, 2015 (5 years ago)

Published:

April 07, 2015 at 21:02:03 CET

Submissions:

1

Project page / code:

Open source:

Yes

Hardware:

2.6 GHz, 1 core

Runtime:

1.4 Hz

Benchmark performance:

Sequence MOTA IDF1 MOTP MT ML FP FN Recall Precision FAF ID Sw. Frag
2D MOT 201523.129.470.934 (4.7)375 (52.0)10,40435,84441.771.11.81,018 (24.4)1,061 (25.5)

Detailed performance:

Sequence MOTA IDF1 MOTP MT ML FP FN Recall Precision FAF ID Sw. Frag
ADL-Rundle-11.030.470.3644,4494,62850.351.38.9136170
ADL-Rundle-318.122.971.8292,7555,35547.363.64.4217140
AVG-TownCentre11.923.070.321583535,87217.978.30.87475
ETH-Crossing22.829.172.91171375324.995.10.188
ETH-Jelmoli43.552.472.99132951,10256.682.90.73751
ETH-Linthescher18.324.077.73146987,12420.294.90.17275
KITTI-1638.839.570.10214286349.385.50.73648
KITTI-1933.841.769.94138872,55252.275.90.8100126
PETS09-S2L246.627.267.6465604,35454.890.41.3238264
TUD-Crossing58.246.670.8323240363.495.60.22632
Venice-118.226.172.9058202,83837.867.81.87472

Raw data: