RGB-D Structural Classification of Guardrails via Learning from Synthetic Data

Kai Göbel, Csaba Beleznai, Alexander Sing, Jürgen Biber, Christian Stefan

2022

Abstract

Vision-based environment perception is a key sensing and analysis modality for mobile robotic platforms. Modern learning concepts allow for interpreting a scene in terms of its objects and their spatial relations. This paper presents a specific analysis pipeline targeting the structural classification of guardrail structures within roadside environments from a mobile platform. Classification implies determining the type label of an observed structure, given a catalog of all possible types. To this end, the proposed concept employs semantic segmentation learned fully in the synthetic domain, and stereo depth data analysis for estimating the metric dimensions of key structural elements. The paper introduces a Blender-based procedural data generation pipeline, targeting to accomplish a narrow sim-to-real gap, allowing to use synthetic training image data to train models valid in the real-world domain. The paper evaluates two semantic segmentation schemes for the part segmentation task, and presents a temporal tracking and propagation concept to aggregate single-frame estimates. Results demonstrate that the proposed analysis framework is well applicable to real scenarios and it can be used as a tool for digitally mapping safety-critical roadside assets.

Paper Citation

in Harvard Style

Göbel K., Beleznai C., Sing A., Biber J. and Stefan C. (2022). RGB-D Structural Classification of Guardrails via Learning from Synthetic Data. In Proceedings of the 14th International Joint Conference on Computational Intelligence - Volume 1: ROBOVIS; ISBN 978-989-758-611-8, SciTePress, pages 445-453. DOI: 10.5220/0011561300003332

in Bibtex Style

@conference{robovis22,
author={Kai Göbel and Csaba Beleznai and Alexander Sing and Jürgen Biber and Christian Stefan},
title={RGB-D Structural Classification of Guardrails via Learning from Synthetic Data},
booktitle={Proceedings of the 14th International Joint Conference on Computational Intelligence - Volume 1: ROBOVIS},
year={2022},
pages={445-453},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011561300003332},
isbn={978-989-758-611-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 14th International Joint Conference on Computational Intelligence - Volume 1: ROBOVIS
TI - RGB-D Structural Classification of Guardrails via Learning from Synthetic Data
SN - 978-989-758-611-8
AU - Göbel K.
AU - Beleznai C.
AU - Sing A.
AU - Biber J.
AU - Stefan C.
PY - 2022
SP - 445
EP - 453
DO - 10.5220/0011561300003332
PB - SciTePress