Garment Particles: A 2D--3D Symmetric Garment Representation for Generation and Editing

2026-05-25

AI Summary

Existing approaches struggle to jointly handle high-level intent-driven garment generation (e.g., from text or images) and the coordinated editing of low-level 2D sewing patterns and 3D geometry. This work proposes Garment Particles—a novel 5D point cloud representation that, for the first time, enables a symmetric and unified encoding of 2D sewing patterns and 3D garment geometry. Built upon a rectified flow framework, the method supports controllable generation of simulation-ready garments from high-level inputs and facilitates multi-dimensional editing. By introducing a Particles-to-Pattern Flow module, the approach efficiently converts point clouds into curvilinear sewing patterns, further enhanced by diffusion-based posterior sampling for high-fidelity synthesis. Experiments demonstrate state-of-the-art performance across multiple datasets, with flexible editing capabilities including interpolation, pattern manipulation, and conditional generation from point clouds or silhouettes.

Abstract

Practical garment design spans two modes: intuitive creation from high-level intent, such as a reference image or text description, and complex low-level editing across 2D sewing patterns and 3D draped geometry, which requires professional training to navigate their complex interdependencies. Yet existing frameworks address only part of this challenge, offering either garment generation from casual inputs or direct editing on sewing patterns. To support both ends of the spectrum, we propose Garment Particles, a 5D point-cloud representation that jointly encodes 2D sewing patterns and 3D geometry. This representation enables Garment Particles Flow (GPF), a rectified flow framework that supports intuitive generation from high-level inputs (text, images, sketches) and various editing operations on 2D sewing patterns and 3D geometries via diffusion posterior sampling. Finally, we introduce Particles-to-Pattern Flow that converts generated garment particles into curved-based patterns for simulation. We validate our model's generation ability on multiple datasets, achieving state-of-the-art garment generation results against competitive baselines. Our model also enables many garment editing scenarios, including garment interpolation, sewing pattern editing, point-cloud- and silhouette-conditioned garment generation. Our project website is at https://garment-particles.github.io .

Problem

• garment generation
• garment editing
• 2D sewing patterns
• 3D geometry
• symmetric representation

Innovation

• Garment Particles
• 2D–3D Symmetric Representation
• Rectified Flow
• Diffusion Posterior Sampling
• Garment Editing