Method

3D-CNN/PMBM [on] [gp] [3D-CNN/PMBM]


Submitted on 27 Jan. 2018 01:51 by
Samuel Scheidegger (Chalmers University of Technology)

Running time:0.01 s
Environment:1 core @ 3.0 Ghz (Matlab)

Method Description:
Output from CNN detector processed by PMBM filter.
Parameters:
threshold=0.9
Latex Bibtex:
@inproceedings{Scheidegger2018,
author = {Samuel Scheidegger and
Joachim Benjaminsson and
Emil Rosenberg and
Amrit Krishnan and
Karl Granström},
title = {Mono-Camera 3D Multi-Object Tracking
Using Deep Learning Detections
and {PMBM} Filtering},
booktitle = {2018 {IEEE} Intelligent Vehicles
Symposium, {IV} 2018, Changshu, Suzhou,
China, June 26-30, 2018},
pages = {433--440},
year = {2018},
crossref = {DBLP:conf/ivs/2018},
url = {https://doi.org/10.1109/IVS.2018.8500454},
doi = {10.1109/IVS.2018.8500454},
timestamp = {Thu, 25 Oct 2018 18:12:58 +0200},
biburl =
{https://dblp.org/rec/bib/conf/ivs/ScheideggerBRKG18},
bibsource = {dblp computer science bibliography,
https://dblp.org}
}

Detailed Results

From all 29 test sequences, our benchmark computes the commonly used tracking metrics CLEARMOT, MT/PT/ML, identity switches, and fragmentations [1,2]. The tables below show all of these metrics.


Benchmark MOTA MOTP MODA MODP
CAR 80.39 % 81.26 % 80.74 % 85.47 %

Benchmark recall precision F1 TP FP FN FAR #objects #trajectories
CAR 85.01 % 96.93 % 90.58 % 31841 1007 5616 9.05 % 36711 1686

Benchmark MT PT ML IDS FRAG
CAR 62.77 % 31.08 % 6.15 % 121 613

This table as LaTeX


[1] K. Bernardin, R. Stiefelhagen: Evaluating Multiple Object Tracking Performance: The CLEAR MOT Metrics. JIVP 2008.
[2] Y. Li, C. Huang, R. Nevatia: Learning to associate: HybridBoosted multi-target tracker for crowded scene. CVPR 2009.


eXTReMe Tracker