Method

3D Multi-Object Tracking in Point Clouds Based on Prediction Confidence-Guided Data Association [CasTrack]


Submitted on 24 Aug. 2024 17:42 by
Mohamed Mostafa (Khalifa University)

Running time:1 s
Environment:1 core @ 2.5 Ghz (Python)

Method Description:
Re-run of the paper "3D Multi-Object Tracking in
Point Clouds Based on Prediction Confidence-Guided
Data Association" CasTrack using their official
code.
Parameters:
N/A
Latex Bibtex:
@ARTICLE{9352500,
author={Wu, Hai and Han, Wenkai and Wen, Chenglu
and Li, Xin and Wang, Cheng},
journal={IEEE Transactions on Intelligent
Transportation Systems},
title={3D Multi-Object Tracking in Point Clouds
Based on Prediction Confidence-Guided Data
Association},
year={2022},
volume={23},
number={6},
pages={5668-5677},
keywords={Three-dimensional
displays;Tracking;Feature extraction;Detectors;Two
dimensional displays;Predictive
models;Acceleration;3D multi-object tracking;point
clouds;data association;object detection and
tracking},
doi={10.1109/TITS.2021.3055616}}

Detailed Results

From all 29 test sequences, our benchmark computes the HOTA tracking metrics (HOTA, DetA, AssA, DetRe, DetPr, AssRe, AssPr, LocA) [1] as well as the CLEARMOT, MT/PT/ML, identity switches, and fragmentation [2,3] metrics. The tables below show all of these metrics.


Benchmark HOTA DetA AssA DetRe DetPr AssRe AssPr LocA
CAR 79.96 % 77.95 % 82.71 % 83.47 % 84.80 % 86.42 % 89.98 % 87.56 %

Benchmark TP FP FN
CAR 32567 1825 1285

Benchmark MOTA MOTP MODA IDSW sMOTA
CAR 90.52 % 86.15 % 90.96 % 150 77.41 %

Benchmark MT rate PT rate ML rate FRAG
CAR 84.46 % 9.54 % 6.00 % 94

Benchmark # Dets # Tracks
CAR 33852 804

This table as LaTeX

This figure as: png pdf

[1] J. Luiten, A. Os̆ep, P. Dendorfer, P. Torr, A. Geiger, L. Leal-Taixé, B. Leibe: HOTA: A Higher Order Metric for Evaluating Multi-object Tracking. IJCV 2020.
[2] K. Bernardin, R. Stiefelhagen: Evaluating Multiple Object Tracking Performance: The CLEAR MOT Metrics. JIVP 2008.
[3] Y. Li, C. Huang, R. Nevatia: Learning to associate: HybridBoosted multi-target tracker for crowded scene. CVPR 2009.


eXTReMe Tracker