SS-IoSR

Abstract

Intraoral 3D scans are fundamental in digital dentistry but often contain geometric artifacts such as holes and non-manifold struc- tures due to acquisition constraints and complex dental anatomy. We introduce SS-IoSR, a self-supervised hybrid framework for intraoral scan repair that combines masked autoencoder-based geometric representa- tion learning with a hybrid explicit–implicit reconstruction strategy. The model operates on localized point cloud patches guided by clinician- selected seed points, enabling anatomically consistent reconstruction of missing regions. A differentiable Poisson-based refinement module fur- ther improves surface continuity and geometric fidelity. Experimental evaluations on public and in-house intraoral scan datasets show that SS-IoSR consistently outperforms state-of-the-art completion methods across multiple metrics, demonstrating improved reconstruction accuracy and higher structural similarity for clinically relevant dental regions.

Method Overview

The proposed framework is trained in two phases.

Phase 1 : We employ a reconstruction-based masked autoencoder (MAE) to reconstruct masked point patches. This pre-training objective encourages the model to learn discriminative local geometric representations, enabling the capture of fine-grained surface details.

Phase 2 : The pre-trained MAE serves as the backbone and is integrated with an implicit refinement module. This module enhances the initially reconstructed point cloud by leveraging the continuous modeling capacity of implicit functions to improve geometric fidelity and surface consistency.

Results

Comparisons with State-of-the-Art point cloud completion methods.

Citing us

For our SS-IoSR work:

@inproceedings{Farhat_2026_MICCAI,
    author    = {Manel Farhat and Achraf Ben-hamadou and Ahmed Rekik and Ons Abida and Oussama Smaoui},
    title     = {SS-IoSR: Self-Supervised Intraoral Scans Repair},
    booktitle = {29th International Conference on Medical Image Computing and Computer Assisted Intervention, {MICCAI} 2026},
    year      = {2026},
}

For the Teeth3DS+ dataset:

@article{ben2022teeth3ds,
    title={{Teeth3Ds+: An Extended Benchmark for Intra-oral 3D Scans Analysis}},
    author={Ben-Hamadou, Achraf and Neifar, Nour and Rekik, Ahmed and Smaoui, Oussama and Bouzguenda, Firas and Pujades, Sergi and Boyer, Edmond and Ladroit, Edouard},
    journal={arXiv preprint arXiv:2210.06094},
    year={2022}
}