TopGen: Learning Structural Layouts and Cross-Fields for Quadrilateral Mesh Generation

2026-03-11

AI Summary

Existing quadrilateral mesh generation methods are often limited by input topology quality and computational efficiency or lack structural awareness and editability. This work proposes TopGen, a novel framework that jointly learns structural layout and cross-field orientation for the first time. By processing triangular meshes through point cloud sampling and a shape encoder, TopGen employs a dual-query decoder to simultaneously predict a geometric skeleton and a consistent cross-field, emulating professional modeling workflows. The method explicitly extracts structural lines and supports downstream editing, significantly outperforming state-of-the-art approaches in both geometric fidelity and topological plausibility, even on high-resolution and non-manifold models. To advance structure-aware quadrangulation research, we also introduce TopGen-220K, a large-scale dataset comprising 220,000 annotated models.

Technology Category

• Machine Learning: Structured Learning
• Natural Language Processing: Generation
• Computer Vision: Generative Adversarial Networks (GANs) for Vision

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Web Mining and Content Analysis: Web data generation and simulation

Abstract

High-quality quadrilateral mesh generation is a fundamental challenge in computer graphics. Traditional optimization-based methods are often constrained by the topological quality of input meshes and suffer from severe efficiency bottlenecks, frequently becoming computationally prohibitive when handling high-resolution models. While emerging learning-based approaches offer greater flexibility, they primarily focus on cross-field prediction, often resulting in the loss of critical structural layouts and a lack of editability. In this paper, we propose TopGen, a robust and efficient learning-based framework that mimics professional manual modeling workflows by simultaneously predicting structural layouts and cross-fields. By processing input triangular meshes through point cloud sampling and a shape encoder, TopGen is inherently robust to non-manifold geometries and low-quality initial topologies. We introduce a dual-query decoder using edge-based and face-based sampling points as queries to perform structural line classification and cross-field regression in parallel. This integrated approach explicitly extracts the geometric skeleton while concurrently capturing orientation fields. Such synergy ensures the preservation of geometric integrity and provides an intuitive, editable foundation for subsequent quadrilateral remeshing. To support this framework, we also introduce a large-scale quadrilateral mesh dataset, TopGen-220K, featuring high-quality paired data comprising raw triangular meshes, structural layouts, cross-fields, and their corresponding quad meshes. Experimental results demonstrate that TopGen significantly outperforms existing state-of-the-art methods in both geometric fidelity and topological edge flow rationality.

Problem

• quadrilateral mesh generation
• structural layouts
• cross-fields
• topological quality
• geometric fidelity

Innovation

• structural layout
• cross-field
• quadrilateral mesh generation
• dual-query decoder
• learning-based remeshing