ML Anthology

Synthetic Data for Defect Segmentation on Complex Metal Surfaces

Juraj Fulir, Lovro Bosnar, Hans Hagen, Petra Gospodnetic
CVPRW 2023 pp. 4424-4434
doi:10.1109/CVPRW59228.2023.00465 /cvprw/2023/fulir2023cvprw-synthetic/

Abstract

Metal defect segmentation poses a great challenge for automated inspection systems due to the complex light reflection from the surface and lack of training data. In this work we introduce a real and synthetic defect segmentation dataset pair for multi-view inspection of a metal clutch part to overcome data shortage. Model pre-training on our synthetic dataset was compared to similar inspection datasets in the literature. Two techniques are presented to increase model training efficiency and prediction coverage in darker areas of the image. Results were collected over three popular segmentation architectures to confirm superior effectiveness of synthetic data and unveil various challenges of multi-view inspection.

CVPRW Semantic Scholar

Cite

Text

Fulir et al. "Synthetic Data for Defect Segmentation on Complex Metal Surfaces." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2023. doi:10.1109/CVPRW59228.2023.00465

Markdown

[Fulir et al. "Synthetic Data for Defect Segmentation on Complex Metal Surfaces." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2023.](https://mlanthology.org/cvprw/2023/fulir2023cvprw-synthetic/) doi:10.1109/CVPRW59228.2023.00465

BibTeX

@inproceedings{fulir2023cvprw-synthetic,
  title     = {{Synthetic Data for Defect Segmentation on Complex Metal Surfaces}},
  author    = {Fulir, Juraj and Bosnar, Lovro and Hagen, Hans and Gospodnetic, Petra},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
  year      = {2023},
  pages     = {4424-4434},
  doi       = {10.1109/CVPRW59228.2023.00465},
  url       = {https://mlanthology.org/cvprw/2023/fulir2023cvprw-synthetic/}
}