Stereo Camera Simulation in Blender

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21730%2F21%3A00346461" target="_blank" >RIV/68407700:21730/21:00346461 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1007/978-3-030-70740-8_13" target="_blank" >https://doi.org/10.1007/978-3-030-70740-8_13</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-030-70740-8_13" target="_blank" >10.1007/978-3-030-70740-8_13</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Stereo Camera Simulation in Blender

Popis výsledku v původním jazyce

The development and implementation of computer vision methods require an appropriate dataset of images for testing. Capturing real images is time-consuming, and it can be difficult in some applications. Furthermore, a precise measurement of distances and geometric transformations between camera positions for particular images is not always possible. In these cases, synthetic data can be beneficially used. Besides, this approach allows precisely setting camera positions and simulating various types of camera movements. The method for generation of stereo camera dataset is presented in this paper. It is built on the 3D creation suite Blender. The method is designed for visual odometry and 3D reconstruction testing, and it simulates a stereo camera movement over the captured 3D model. The images are directly inputted to the tested system. Together with the known ground truth of camera positions, the synthetic data allow testing particular steps of the system. The proposed method was used for testing 3D reconstruction of a camera-based car undercarriages scanner.

Název v anglickém jazyce

Stereo Camera Simulation in Blender

Popis výsledku anglicky

The development and implementation of computer vision methods require an appropriate dataset of images for testing. Capturing real images is time-consuming, and it can be difficult in some applications. Furthermore, a precise measurement of distances and geometric transformations between camera positions for particular images is not always possible. In these cases, synthetic data can be beneficially used. Besides, this approach allows precisely setting camera positions and simulating various types of camera movements. The method for generation of stereo camera dataset is presented in this paper. It is built on the 3D creation suite Blender. The method is designed for visual odometry and 3D reconstruction testing, and it simulates a stereo camera movement over the captured 3D model. The images are directly inputted to the tested system. Together with the known ground truth of camera positions, the synthetic data allow testing particular steps of the system. The proposed method was used for testing 3D reconstruction of a camera-based car undercarriages scanner.

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/VI20172020080" target="_blank" >VI20172020080: Kassandra - mnohokamerový bezpečnostní skener podvozků vozidel</a><br>

Návaznosti

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

Rok uplatnění

2021

Kód důvěrnosti údajů

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

Název statě ve sborníku

Modelling and Simulation for Autonomous Systems

ISBN

978-3-030-70739-2

ISSN

0302-9743

e-ISSN

1611-3349

Počet stran výsledku

11

Strana od-do

206-216

Název nakladatele

Springer, Cham

Místo vydání

—

Místo konání akce

Praha

Datum konání akce

21. 10. 2020

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000763018100013