Method

Complexer-YOLO: Real-Time 3D Object Detection and Tracking on Semantic Point Clouds [la] [on] [gp] [Complexer-YOLO]


Submitted on 16 Nov. 2018 13:43 by
Martin Simon (Valeo Schalter und Sensoren GmbH)

Running time:0.01 a
Environment:GPU @ 3.5 Ghz (C/C++)

Method Description:
Parameters:
Latex Bibtex:
@inproceedings{Simon_2019_CVPR_Workshops,
author = {Simon, Martin and Amende, Karl and
Kraus, Andrea and Honer, Jens and Samann, Timo
and Kaulbersch, Hauke and Milz, Stefan and
Michael Gross, Horst},
title = {Complexer-YOLO: Real-Time 3D Object
Detection and Tracking on Semantic Point
Clouds},
booktitle = {The IEEE Conference on Computer
Vision and Pattern Recognition (CVPR)
Workshops},
month = {June},
year = {2019}}

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 75.70 % 78.46 % 79.15 % 82.83 %
PEDESTRIAN 16.46 % 62.69 % 18.73 % 89.09 %

Benchmark recall precision F1 TP FP FN FAR #objects #trajectories
CAR 85.32 % 95.18 % 89.98 % 32215 1631 5541 14.66 % 37580 3239
PEDESTRIAN 35.63 % 68.31 % 46.83 % 8285 3843 14970 34.55 % 16707 2300

Benchmark MT PT ML IDS FRAG
CAR 58.00 % 36.92 % 5.08 % 1186 2092
PEDESTRIAN 2.41 % 59.45 % 38.14 % 527 1636

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