HAM_SADF: Historical Appearance Matching and Scene Adaptive Detection Filtering

TUD-Crossing


Benchmark:

Short name:

HAM_SADF

Detector:

Public

Description:

https://github.com/yyc9268/Deep-TAMA

Reference:

Y. Yoon, A. Boragule, Y. Song, K. Yoon, M. Jeon. Online Multi-Object Tracking with Historical Appearance Matching and Scene Adaptive Detection Filtering. In IEEE AVSS, 2018.

Last submitted:

June 13, 2018 (2 years ago)

Published:

June 04, 2018 at 07:12:57 CET

Submissions:

3

Open source:

Yes

Hardware:

3.7 GHZ, 4 Core, TITAN X

Runtime:

18.7 Hz

Benchmark performance:

Sequence MOTA IDF1 MOTP MT ML FP FN Recall Precision FAF ID Sw. Frag
2D MOT 201525.237.871.441 (5.7)420 (58.3)7,33038,27537.776.01.3357 (9.5)745 (19.8)

Detailed performance:

Sequence MOTA IDF1 MOTP MT ML FP FN Recall Precision FAF ID Sw. Frag
ADL-Rundle-111.938.071.1683,4494,69549.557.26.956121
ADL-Rundle-333.242.073.14111,2595,46846.278.92.06061
AVG-TownCentre14.526.570.051582425,84118.384.40.52778
ETH-Crossing26.141.475.21182271828.492.80.112
ETH-Jelmoli37.957.074.17211611,40844.587.50.4730
ETH-Linthescher15.523.673.56161847,44816.694.60.11135
KITTI-1638.347.873.1048894444.589.60.41833
KITTI-1926.043.866.94217463,16040.974.50.748142
PETS09-S2L239.237.669.83105495,21645.989.01.3101189
TUD-Crossing58.968.174.8427237266.291.00.4913
Venice-119.333.172.1166583,00534.170.31.51941

Raw data: