MidSurfNet: Learnable Face Pairing and Interference Implicit Fields for Generalized Mid-surface Abstraction

2026-06-01

AI Summary

Existing medial surface extraction methods based on face pairing rely on handcrafted geometric heuristics, limiting their effectiveness on complex CAD models featuring multi-thickness walls, self-matching regions, or non-central offsets. To address this, this work proposes MidSurfNet, which introduces a learnable neural face-pairing module to automatically capture both geometric and topological features. The method further enables flexible offset control by modeling the medial surface through a perturbed implicit field. Trained and evaluated on over 1,500 manually annotated industrial CAD models, MidSurfNet achieves a face-pairing accuracy of 87.32% and demonstrates significantly improved medial surface completion rates—61.90% for multi-thickness cases and 52.94% for self-matching scenarios—outperforming existing rule-based approaches.

Abstract

Mid-surface abstraction is essential for finite element analysis of thin-walled CAD models. Existing face pairing-based methods rely on handcrafted geometric heuristics, yet real-world industrial models frequently exhibit multi-wall-thickness regions, self-matching face configurations, and demand for non-center offset surfaces--scenarios where rule-based approaches consistently fail. We present MidSurfNet, a learning-augmented framework that addresses these limitations through two novel components: (1) a neural face pairing module that learns to predict face pair confidence from geometric and topological features, handling complex pairing scenarios beyond rule-based methods; and (2) an interference implicit field that represents mid-surfaces as the interference of two signed distance functions, enabling generalized offset control for flexible positioning in downstream CAE/FEA-oriented workflows. We construct a large-scale mid-surface dataset containing over 1,500 manually annotated CAD models. Experiments demonstrate that MidSurfNet achieves 87.32% face pairing accuracy and successfully handles multi-wall-thickness (61.90% completion) and self-matching (52.94% completion) scenarios that confound all existing methods. Furthermore, MidSurfNet provides a learning-based approach to generalized mid-surface abstraction with arbitrary offset control for CAE-oriented applications.

Problem

• mid-surface abstraction
• face pairing
• multi-wall-thickness
• self-matching
• offset control

Innovation

• learnable face pairing
• interference implicit field
• mid-surface abstraction
• signed distance function
• CAE-oriented offset control