GeoT: Geometry-guided Instance-dependent Transition Matrix for Semi-supervised Tooth Point Cloud Segmentation

TL;DR

GeoT addresses semi-supervised tooth point cloud segmentation by modeling pseudo-label noise with an instance-dependent transition matrix (IDTM) guided by geometry. It introduces point-level geometric regularization (PLGR) and class-level geometric smoothing (CLGS) to constrain IDTM estimation, fusing per-point and class priors into a unified objective. The approach shows state-of-the-art performance on Teeth3DS and a private dataset, achieving results close to fully supervised learning with only 20% labeled data. This work significantly improves the利用 of unlabeled intra-oral scan data for accurate orthodontic segmentation, reducing labeling burden and enhancing clinical applicability.

Abstract

Achieving meticulous segmentation of tooth point clouds from intra-oral scans stands as an indispensable prerequisite for various orthodontic applications. Given the labor-intensive nature of dental annotation, a significant amount of data remains unlabeled, driving increasing interest in semi-supervised approaches. One primary challenge of existing semi-supervised medical segmentation methods lies in noisy pseudo labels generated for unlabeled data. To address this challenge, we propose GeoT, the first framework that employs instance-dependent transition matrix (IDTM) to explicitly model noise in pseudo labels for semi-supervised dental segmentation. Specifically, to handle the extensive solution space of IDTM arising from tens of thousands of dental points, we introduce tooth geometric priors through two key components: point-level geometric regularization (PLGR) to enhance consistency between point adjacency relationships in 3D and IDTM spaces, and class-level geometric smoothing (CLGS) to leverage the fixed spatial distribution of tooth categories for optimal IDTM estimation. Extensive experiments performed on the public Teeth3DS dataset and private dataset demonstrate that our method can make full utilization of unlabeled data to facilitate segmentation, achieving performance comparable to fully supervised methods with only $20\%$ of the labeled data.

Figures (4)

• Figure 1: Overview of the proposed GeoT framework. The top portion presents the flowchart of the method, which encompasses the proposed point-level geometric regularization and class-level geometric smoothing. The lower portion illustrates the impact of these two modules on the noise transition matrix. Point 1 and 2 are the same class, point 3 is different. The $4\times4$ grid represents NTM with 4 classes, where a darker color indicates a higher transition probability. Categories 1 and 2, 2 and 3, as well as 3 and 4 are spatially positioned close to each other. PLGR aligns NTMs of the same class and separates different ones. CLGS corrects NTMs violating distribution law.
• Figure 2: Visualization of the segmentation results of different methods. The first four rows are the lower jaws, and the last two rows are the upper jaws.
• Figure 3: Experiment resuls with different weighting factors $\lambda$ for $T^{F}_{k}$ in different splits.
• Figure 4: Experiment resuls with different weights $\beta$ for $\mathcal{L}_{m}$ in different splits.