MPNTrack: Learning a Neural Solver for Multiple Object Tracking

MOT16-01


Benchmark:

Short name:

MPNTrack

Detector:

Public

Description:

Graphs offer a natural way to formulate Multiple Object Tracking (MOT) within the tracking-by-detection paradigm. However, they also introduce a major challenge for learning methods, as defining a model that can operate on such a structured domain is not trivial. As a consequence, most learning-based work has been devoted to learning better features for MOT, and then using these with well-established optimization frameworks. In this work, we exploit the classical network flow formulation of MOT to define a fully differentiable framework based on Message Passing Networks (MPNs). By operating directly on the graph domain, our method can reason globally over an entire set of detections and predict final solutions. Hence, we show that learning in MOT does not need to be restricted to feature extraction, but it can also be applied to the data association step. We show a significant improvement in both MOTA and IDF1 on three publicly available benchmarks.

Reference:

G. Brasó, L. Leal-Taixé. Learning a Neural Solver for Multiple Object Tracking. In CVPR, 2020.

Last submitted:

April 16, 2020 (2 months ago)

Published:

April 17, 2020 at 10:42:16 CET

Submissions:

1

Open source:

Yes

Hardware:

NVIDIA Quadro P5000

Runtime:

6.5 Hz

Benchmark performance:

Sequence MOTA IDF1 MOTP MT ML FP FN Recall Precision FAF ID Sw. Frag
MOT1658.661.778.9207 (27.3)258 (34.0)4,94970,25261.595.80.8354 (5.8)684 (11.1)

Detailed performance:

Sequence MOTA IDF1 MOTP MT ML FP FN Recall Precision FAF ID Sw. Frag
MOT16-0150.263.876.9810643,12151.298.10.1214
MOT16-0369.267.478.864181,90030,15871.297.51.3116209
MOT16-0656.059.479.477829224,11864.388.90.838101
MOT16-0746.852.077.98125508,08050.593.71.156119
MOT16-0840.048.681.817255229,48443.393.30.84464
MOT16-1247.959.481.420363823,92552.792.00.41636
MOT16-1434.846.276.4137560911,36638.592.10.882141

Raw data:


MPNTrack