Bootstrapped Learning for Car Detection in Planar Lidars

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F22%3A00361168" target="_blank" >RIV/68407700:21230/22:00361168 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1145/3477314.3507312" target="_blank" >https://doi.org/10.1145/3477314.3507312</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1145/3477314.3507312" target="_blank" >10.1145/3477314.3507312</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Bootstrapped Learning for Car Detection in Planar Lidars

Popis výsledku v původním jazyce

We present a proof-of-concept method for using bootstrapped learning for car detection in lidar scans using neural networks. We transfer knowledge from a traditional hand-engineered clustering and geometry-based detection technique to deep-learning-based methods. The geometry-based method automatically annotates laserscans from a vehicle travelling around a static car park over a long period of time. We use these annotations to automatically train the deep-learning neural network and evaluate and compare this method against the original geometrical method in various weather conditions. Furthermore, by using temporal filters, we can find situations where the original method was struggling or giving intermittent detections and still automatically annotate these frames and use them as part of the training process. Our evaluation indicates an increased detection accuracy and robustness as sensing conditions deteriorate compared to the method from which trained the neural network.

Název v anglickém jazyce

Bootstrapped Learning for Car Detection in Planar Lidars

Popis výsledku anglicky

We present a proof-of-concept method for using bootstrapped learning for car detection in lidar scans using neural networks. We transfer knowledge from a traditional hand-engineered clustering and geometry-based detection technique to deep-learning-based methods. The geometry-based method automatically annotates laserscans from a vehicle travelling around a static car park over a long period of time. We use these annotations to automatically train the deep-learning neural network and evaluate and compare this method against the original geometrical method in various weather conditions. Furthermore, by using temporal filters, we can find situations where the original method was struggling or giving intermittent detections and still automatically annotate these frames and use them as part of the training process. Our evaluation indicates an increased detection accuracy and robustness as sensing conditions deteriorate compared to the method from which trained the neural network.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

<a href="/cs/project/GC20-27034J" target="_blank" >GC20-27034J: Rozšíření prostorových modelů explicitní representací času pro dlouhodobou autonomii mobilních robotů</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2022

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Proceedings of the 37th ACM/SIGAPP Symposium on Applied Computing

ISBN

978-1-4503-8713-2

ISSN

—

e-ISSN

—

Počet stran výsledku

8

Strana od-do

758-765

Název nakladatele

ACM

Místo vydání

New York

Místo konání akce

Virtual

Datum konání akce

25. 4. 2022

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—