TCATSeg: A Tooth Center-Wise Attention Network for 3D Dental Model Semantic Segmentation

2026-03-17

AI Summary

Existing methods for semantic segmentation of 3D dental models often struggle to accurately distinguish adjacent teeth that exhibit similar morphologies and are tightly arranged, primarily due to their neglect of global contextual information. To address this limitation, this work proposes TCATSeg, a novel framework that introduces, for the first time, a tooth-center-oriented attention mechanism and constructs physically meaningful sparse superpoints to model global semantic dependencies among teeth. By effectively integrating local geometric features with global contextual cues, TCATSeg achieves significantly improved segmentation accuracy over state-of-the-art methods on a newly introduced dataset comprising 400 pre-orthodontic dental models.

Technology Category

• Computer Vision: Segmentation
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Accurate semantic segmentation of 3D dental models is essential for digital dentistry applications such as orthodontics and dental implants. However, due to complex tooth arrangements and similarities in shape among adjacent teeth, existing methods struggle with accurate segmentation, because they often focus on local geometry while neglecting global contextual information. To address this, we propose TCATSeg, a novel framework that combines local geometric features with global semantic context. We introduce a set of sparse yet physically meaningful superpoints to capture global semantic relationships and enhance segmentation accuracy. Additionally, we present a new dataset of 400 dental models, including pre-orthodontic samples, to evaluate the generalization of our method. Extensive experiments demonstrate that TCATSeg outperforms state-of-the-art approaches.

Problem

• 3D dental model
• semantic segmentation
• tooth segmentation
• global context
• digital dentistry

Innovation

• tooth center-wise attention
• 3D dental segmentation
• superpoints
• global semantic context
• digital dentistry