Semi-supervised Latent Disentangled Diffusion Model for Textile Pattern Generation

2026-03-17

AI Summary

This work addresses the limitations of existing image-to-image translation models in textile pattern generation, which often suffer from detail distortion and low fidelity due to feature entanglement and non-rigid texture deformations. To overcome these challenges, the authors propose SLDDM-TPG, a two-stage framework that first employs a Latent Disentanglement Network (LDN) to construct a multidimensional, disentangled garment feature space, effectively separating intricate patterns from deformation-related features. Subsequently, a semi-supervised Latent Diffusion Model (S-LDM) combined with a fine-grained alignment strategy enables high-fidelity pattern synthesis. This study presents the first integration of latent disentanglement with semi-supervised diffusion mechanisms for textile pattern generation, achieving a 4.1 reduction in FID and a 0.116 improvement in SSIM on the CTP-HD dataset, while also demonstrating strong generalization performance on VITON-HD.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Generation

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Textile pattern generation (TPG) aims to synthesize fine-grained textile pattern images based on given clothing images. Although previous studies have not explicitly investigated TPG, existing image-to-image models appear to be natural candidates for this task. However, when applied directly, these methods often produce unfaithful results, failing to preserve fine-grained details due to feature confusion between complex textile patterns and the inherent non-rigid texture distortions in clothing images. In this paper, we propose a novel method, SLDDM-TPG, for faithful and high-fidelity TPG. Our method consists of two stages: (1) a latent disentangled network (LDN) that resolves feature confusion in clothing representations and constructs a multi-dimensional, independent clothing feature space; and (2) a semi-supervised latent diffusion model (S-LDM), which receives guidance signals from LDN and generates faithful results through semi-supervised diffusion training, combined with our designed fine-grained alignment strategy. Extensive evaluations show that SLDDM-TPG reduces FID by 4.1 and improves SSIM by up to 0.116 on our CTP-HD dataset, and also demonstrate good generalization on the VITON-HD dataset.

Problem

• textile pattern generation
• feature confusion
• non-rigid texture distortion
• fine-grained detail preservation
• image-to-image translation

Innovation

• latent disentanglement
• semi-supervised diffusion model
• textile pattern generation
• feature confusion resolution
• fine-grained alignment