Rendering Synthetic Defects for Learning‐Based Industrial Inspection
Computer Graphics Forum · 12.08.2026
DOI: 10.1111/cgf.70550 ↗Kurz gefasst
Fachartikel in der Zeitschrift Computer Graphics Forum, online erschienen am 12.08.2026. Verfasst von Runzhou Mao und 5 weiteren Autor:innen. Die Arbeit nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Vienna Science and Technology Fund.
Abstract (Originalsprache)
Computer vision increasingly uses synthetic data from physically based rendering to supplement limited real‐world datasets. In industrial inspection, defect data is scarce and the rendering pipeline is explicitly controlled, and synthetic defect generation therefore becomes a dataset design problem. In this setting, building a dataset means choosing points in a rendering parameter space: defect shape, material, illumination, viewpoint, and sampling define the training distribution, but their effect on downstream learning is often hard to judge from images alone. We therefore study how these factors change defect features, where their relation to downstream learning is easier to inspect. Our results show that the rendering factors do not matter equally: defect shape, material, illumination, and viewpoint often affect downstream behavior much more than the number of samples per pixel. Synthetic subsets that transfer better downstream tend to stay close to real defect features, cover the observed defect modes, and stay separated from the defect‐free (OK) region. Based on these observations, we build a simple feature‐space screening heuristic for selecting subsets from large candidate pools. The selected subsets often outperform matched random selection for downstream segmentation on real data.
Förderung
- Vienna Science and Technology Fund
Zitieren
APA
Mao, R., Sakai, H., Freude, C., Garth, C., Gospodnetić, P., & Fulir, J. (2026). Rendering Synthetic Defects for Learning‐Based Industrial Inspection. Computer Graphics Forum. https://doi.org/10.1111/cgf.70550
Vancouver
Mao R, Sakai H, Freude C, Garth C, Gospodnetić P, Fulir J. Rendering Synthetic Defects for Learning‐Based Industrial Inspection. Computer Graphics Forum. 2026. doi:10.1111/cgf.70550
BibTeX
@article{Mao2026,
author = {Mao, Runzhou and Sakai, Hiroyuki and Freude, Christian and Garth, Christoph and Gospodnetić, Petra and Fulir, Juraj},
title = {Rendering Synthetic Defects for Learning‐Based Industrial Inspection},
journal = {Computer Graphics Forum},
year = {2026},
doi = {10.1111/cgf.70550}
}
EUVIMED