AnySurf: Any Surface Generation with Directed Edge

2026-05-22

AI Summary

Existing 3D generation methods struggle to uniformly handle open, closed, and mixed surfaces, often producing normal inconsistencies and topological defects in industrial applications such as garment modeling. This work proposes AnySurf, a unified framework that leverages a directed-edge-enhanced Flexible Dual Grid (FDG-D) representation to explicitly encode surface orientation, thereby eliminating bilayer artifacts and normal ambiguities. A lightweight DE-Adapter module, adding only 1% more parameters, is introduced alongside a ROS-FT post-training strategy to enable high-quality generation of arbitrary surfaces. Evaluated on the newly curated industrial-scale Outfit3D dataset, AnySurf significantly outperforms existing approaches, yielding meshes with superior geometric fidelity and correct face orientations, and demonstrates enhanced practicality in downstream tasks including rendering, physical simulation, and geometry editing.

Abstract

Open surface components prevail in real industrial 3D content and support rendering, physical simulation and geometric editing. Garments serve as a typical open surface type, with numerous existing generation methods leveraging sewing patterns to generate 2D panels and stitch them into 3D shapes. Such domain-specific designs lack scalability and cannot generalize to shoes and accessories. Common field-based 3D generators prioritize watertight meshes and tend to create flawed double-layer structures on open surfaces. Though Trellis2 adopts field-free representation, its open surface results still contain normal and topology errors. We present AnySurf, a unified framework generating open, closed and hybrid 3D surfaces with accurate face orientation. Built on directed-edge enhanced Flexible Dual Grid (FDG-D), our representation retains normal direction information via oriented grid edges. We also propose ROS-FT post-training and a lightweight DE-Adapter with merely 1% extra parameters, facilitating directed edge learning while preserving original generation performance. We further construct Outfit3D dataset containing industrial garments and closed accessories. Our work transforms garment modeling into a universal 3D generation task. Experimental results demonstrate superior mesh quality and better practicality for downstream applications.

Problem

• open surface
• 3D generation
• mesh quality
• normal orientation
• industrial 3D content

Innovation

• directed edge
• open surface generation
• Flexible Dual Grid
• face orientation
• universal 3D generation