Historical Appearance Matching and Scene Adaptive Detection Filtering

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

Short name:

HAM_SADF

Benchmark:

Description:

n/a

Hardware:

3.7 GHZ, 4 Core, TITAN X

Detector:

Public

Processing:

Online

Last submitted:

June 13, 2018 (6 months ago)

Published:

June 04, 2018 at 07:12:57 CET

Submissions:

3

Open source:

No

Project page / code:

n/a

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.

Benchmark performance:

MOTAMOTPFAFMTMLFPFNID Sw.FragSpecificationsDetector
25.271.41.35.7 % 58.3 % 7,33038,2753577453.7 GHZ, 4 Core, TITAN XPublic
IDF1ID PrecisionID Recall
37.857.028.3

Detailed performance:

Sequence MOTA IDF1 MOTP FAF GT MT ML FP FN ID Sw Frag
TUD-Crossing58.968.174.80.41330.8 % 15.4 % 72372913
PETS09-S2L239.237.669.81.3427.1 % 23.8 % 5495,216101189
ETH-Jelmoli37.957.074.10.44515.6 % 46.7 % 1611,408730
ETH-Linthescher15.523.673.50.11973.0 % 81.7 % 847,4481135
ETH-Crossing26.141.475.20.1263.8 % 69.2 % 2271812
AVG-TownCentre14.526.570.00.52262.2 % 69.9 % 2425,8412778
ADL-Rundle-111.938.071.16.93218.8 % 25.0 % 3,4494,69556121
ADL-Rundle-333.242.073.12.0449.1 % 25.0 % 1,2595,4686061
KITTI-1638.347.873.10.4170.0 % 23.5 % 889441833
KITTI-1926.043.866.90.7626.5 % 33.9 % 7463,16048142
Venice-119.333.172.11.5175.9 % 35.3 % 6583,0051941

Raw data:


HAM_SADF
YT