EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Explaining 3D Semantic Segmentation Through Generative AI ‐Based Counterfactuals

Dzemail Rozajac, Niko Lukač, Stefan Schweng, Christoph Gollob, Arne Nothdurft, Karl Stampfer, Javier Del Ser, Andreas Holzinger

Expert Systems · 2025-12 · 42 (12)

DOI: 10.1111/exsy.70163 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Expert Systems, online erschienen am 11.11.2025. Verfasst von Dzemail Rozajac und 7 weiteren Autor:innen. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 52 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Austrian Science Fund, Eusko Jaurlaritza.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Interpreting the predictions of deep learning models on 3D point cloud data is an important challenge for safety‐critical domains such as autonomous driving, robotics and geospatial analysis. Existing counterfactual explainability methods often struggle with the sparsity and unordered nature of 3D point clouds. To address this, we introduce a generative framework for counterfactual explanations in 3D semantic segmentation models. Our approach leverages autoencoder‐based latent representations, combined with UMAP embeddings and Delaunay triangulation, to construct a graph that enables geodesic path search between semantic classes. Candidate counterfactuals are generated by interpolating latent vectors along these paths and decoding into plausible point clouds, while semantic plausibility is guided by the predictions of a 3D semantic segmentation model. We evaluate the framework on ShapeNet objects, demonstrating that semantically related classes yield realistic counterfactuals with minimal geometric change, whereas unrelated classes expose sharp decision boundaries and reduced plausibility. Quantitative results confirm that the method balances defined interpretability metrics, producing counterfactuals that are both interpretable and geometrically consistent. Overall, our work demonstrates that generative counterfactuals in latent space provide a promising alternative to input‐level perturbations.

Förderung

  • Austrian Science Fund
  • Eusko Jaurlaritza

Zitieren

APA

Rozajac, D., Lukač, N., Schweng, S., Gollob, C., Nothdurft, A., Stampfer, K., Ser, J. D., & Holzinger, A. (2025). Explaining 3D Semantic Segmentation Through Generative AI ‐Based Counterfactuals. Expert Systems, 42(12). https://doi.org/10.1111/exsy.70163

Vancouver

Rozajac D, Lukač N, Schweng S, Gollob C, Nothdurft A, Stampfer K, et al. Explaining 3D Semantic Segmentation Through Generative AI ‐Based Counterfactuals. Expert Systems. 2025;42(12). doi:10.1111/exsy.70163

BibTeX
@article{Rozajac2025,
  author = {Rozajac, Dzemail and Lukač, Niko and Schweng, Stefan and Gollob, Christoph and Nothdurft, Arne and Stampfer, Karl and Ser, Javier Del and Holzinger, Andreas},
  title = {Explaining 3D Semantic Segmentation Through Generative AI ‐Based Counterfactuals},
  journal = {Expert Systems},
  year = {2025},
  volume = {42},
  number = {12},
  doi = {10.1111/exsy.70163}
}
Kostenlos starten →