In the fast-paced world of fashion, the gap between creative ideation and technical production has long been a bottleneck. Style3D, a company specializing in 3D and AI technologies for the fashion industry, is aiming to close this gap by linking generative AI with structured product data, enabling a more connected workflow from design to production.
At the heart of this innovation is GarmageNet, a framework developed by Style3D Research in collaboration with Zhejiang University, Shanghai Jiao Tong University, and Zhejiang Sci-Tech University. Published in ACM Transactions on Graphics and presented at SIGGRAPH Asia 2025, GarmageNet addresses a central challenge in fashion AI: bridging 2D patterns with 3D garment forms. The framework captures pattern outlines, 3D geometry, and sewing information in a unified representation, allowing inputs such as text, sketches, images, patterns, or point clouds to be transformed into structured assets. It can reconstruct sewing relationships and initialize physical simulations, facilitating design exploration, digital sampling, and editing.
According to the research, GarmageNet is supported by GarmageSet, a dataset comprising 14,801 professionally created garments. In evaluations, the system demonstrated high accuracy: GarmageJigsaw achieved 99.16% precision and 97.13% recall in sewing-point identification, and GarmageNet achieved a 91.41% simulation-initialization success rate on 150 complex patterns, with an inference time of approximately eight seconds per garment under test conditions.
This research reflects Style3D’s broader commitment to advancing fashion technology. The company has published over 30 papers at major conferences including SIGGRAPH, CVPR, and NeurIPS, and holds 106 granted patents worldwide, including 58 invention patents and 12 granted in Europe and the United States, with 155 more pending. Style3D also operates an overseas R&D center in Germany.
The practical impact of these technologies is already being felt in the industry. German menswear brand OLYMP uses Style3D and Assyst to integrate 2D patterns, 3D simulation, visualization, and digital showrooms. According to the company, this integration has reduced style development from weeks to days, cut the number of physical samples, and improved communication across design, sales, and production.
Style3D’s vision extends beyond mere content generation. As CEO Eric Liu told Just Style, “The most important development is the agent. We use agents to link different skills.” By leveraging AI agents to coordinate product data, 3D design, simulation, and production preparation, Style3D aims to create a seamless workflow where human teams retain creative and operational control. This shift from isolated AI tools to integrated systems marks a significant step forward for the fashion industry, promising increased efficiency, reduced waste, and faster time-to-market.
