PatternGSL: A Structured Specification Language for Template-Free and Simulation-Ready 3D Garments

2026-06-23

AI Summary

Single-image 3D garment reconstruction faces a fundamental trade-off between geometric fidelity and the preservation of simulatable sewing structures: template-free methods lack explicit sewing information, while template-based approaches are constrained by predefined topologies. This work proposes PatternGSL—a template-free, learnable structured language for garment representation—that uniquely treats sewing structure as a first-class objective in generative modeling. Our method directly predicts a complete sewing graph from a single image, including pattern piece boundaries, parameterized seams, and stitching topology, and employs a lightweight deterministic decoder to produce simulation-ready 3D garments. We introduce PatternGSLData, the first large-scale dataset of image-to-GSL pairs, and demonstrate significant improvements over existing methods in sewing graph accuracy, structural completeness, simulation reliability, and pattern editability, enabling end-to-end generation of physically simulatable 3D garments.

Abstract

Reconstructing realistic, physically plausible garments from a single image remains a fundamental challenge. Template-free methods capture surface geometry but lack explicit sewing structure for simulation; while programmatic systems are simulation-ready but constrained by predefined templates. This reveals a fundamental representation gap between geometric reconstruction and structured garment construction. We present PatternGSL, a structured garment representation in the form of a template-free and learnable specification language that encodes complete sewing patterns, including panel boundaries, parameterized seams, and explicit stitch topology, in a compact and standardized form. PatternGSL preserves the physical rigor of pattern-based models while removing template dependence, elevating sewing structure as a first-class target for generative modeling. We further propose a vision-language framework that predicts PatternGSL specifications directly from a single image and decodes them into garments using lightweight deterministic validity handling, without optimization-based refinement or manual cleanup. In addition, we introduce PatternGSLData, the first large-scale image-to-GSL paired dataset comprising 300K samples with complete sewing pattern annotations, enabling supervised VLM training for structured garment reconstruction. Experiments demonstrate improved pattern accuracy over prior baselines, explicit sewing-structure recovery, reliable cloth simulation, and pattern-level editing through the same deterministic decoding pipeline. Code and data-processing scripts will be released at https://github.com/PatternGSL/PatternGSL.

Problem

• garment reconstruction
• sewing structure
• template-free
• 3D garments
• single-image

Innovation

• PatternGSL
• template-free garment representation
• sewing structure
• simulation-ready 3D garments
• vision-language modeling