TSegAgent: Zero-Shot Tooth Segmentation via Geometry-Aware Vision-Language Agents

2026-03-20

AI Summary

This work addresses the limitations of conventional dental segmentation methods, which heavily rely on large annotated datasets and exhibit poor generalization. To overcome these challenges, the authors propose a zero-shot geometric reasoning framework that reframes dental segmentation as a geometry-aware understanding task, eliminating the need for task-specific training. By integrating general-purpose vision-language foundation models with geometric priors derived from dental anatomical structures, the method leverages multi-view visual abstraction, dental arch modeling, and volumetric spatial reasoning to achieve high-precision instance-level segmentation on unseen intraoral scans. This study represents the first application of geometry-informed zero-shot inference to dental image analysis, significantly reducing annotation dependency while enhancing cross-scanner generalization and computational efficiency.

Technology Category

• Computer Vision: Segmentation
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Deep Neural Architectures and Foundation Models

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Automatic tooth segmentation and identification from intra-oral scanned 3D models are fundamental problems in digital dentistry, yet most existing approaches rely on task-specific 3D neural networks trained with densely annotated datasets, resulting in high annotation cost and limited generalization to scans from unseen sources. Thus, we propose TSegAgent, which addresses these challenges by reformulating dental analysis as a zero-shot geometric reasoning problem rather than a purely data-driven recognition task. The key idea is to combine the representational capacity of general-purpose foundation models with explicit geometric inductive biases derived from dental anatomy. Instead of learning dental-specific features, the proposed framework leverages multi-view visual abstraction and geometry-grounded reasoning to infer tooth instances and identities without task-specific training. By explicitly encoding structural constraints such as dental arch organization and volumetric relationships, the method reduces uncertainty in ambiguous cases and mitigates overfitting to particular shape distributions. Experimental results demonstrate that this reasoning-oriented formulation enables accurate and reliable tooth segmentation and identification with low computational and annotation cost, while exhibiting strong generalization across diverse and previously unseen dental scans.

Problem

• tooth segmentation
• zero-shot learning
• 3D dental models
• generalization
• annotation cost

Innovation

• zero-shot segmentation
• geometry-aware reasoning
• vision-language agent
• dental 3D modeling
• inductive bias