RNN_LSTM: RNN Tracker


Video not available.

Benchmark:

Short name:

RNN_LSTM

Detector:

Public

Description:

We present a novel approach to online multi-target tracking based on
recurrent neural networks (RNNs). Tracking multiple objects in
real-world scenes involves many challenges, including a) an a-priori
unknown and time-varying number of targets, b) a continuous state
estimation of all present targets, and c) a discrete combinatorial
problem of data association. Most previous methods involve complex
models that require tedious tuning of parameters. Here, we propose for
the first time, a full end-to-end learning approach for online
multi-target tracking based on deep learning. Existing deep learning
methods are not designed for the above challenges and cannot be
trivially applied to the task. Our solution addresses all of the above
points in a principled way. Experiments on both synthetic and real data
show competitive results obtained at 300 Hz on a standard CPU, and pave
the way towards future research in this direction.

Reference:

A. Milan, S. Rezatofighi, A. Dick, I. Reid, K. Schindler. Online Multi-Target Tracking using Recurrent Neural Networks. In AAAI, 2017.

Last submitted:

March 15, 2016 (4 years ago)

Published:

March 15, 2016 at 08:24:33 CET

Submissions:

2

Open source:

Yes

Hardware:

3 GHz, 1 CPU

Runtime:

165.2 Hz

Benchmark performance:

Sequence MOTA IDF1 MOTP MT ML FP FN Recall Precision FAF ID Sw. Frag
2D MOT 201519.017.171.040 (5.5)329 (45.6)11,57836,70640.368.12.01,490 (37.0)2,081 (51.7)

Detailed performance:

Sequence MOTA IDF1 MOTP MT ML FP FN Recall Precision FAF ID Sw. Frag
ADL-Rundle-10.00.00.000000.00.00.000
ADL-Rundle-30.00.00.000000.00.00.000
AVG-TownCentre0.00.00.000000.00.00.000
ETH-Crossing0.00.00.000000.00.00.000
ETH-Jelmoli0.00.00.000000.00.00.000
ETH-Linthescher0.00.00.000000.00.00.000
KITTI-160.00.00.000000.00.00.000
KITTI-190.00.00.000000.00.00.000
PETS09-S2L20.00.00.000000.00.00.000
TUD-Crossing0.00.00.000000.00.00.000
Venice-10.00.00.000000.00.00.000

Raw data: