The AI fashion revolution is here – and it could change how your clothes are made
From sketches and patterns to 3D garments and production, Style3D is pushing fashion AI beyond image generation - and into the whole design process.
Fashion AI is moving beyond simply creating eye-catching images.
Style3D is working on technology designed to connect generative AI, 3D technology and structured product data, allowing fashion teams to take an idea from the earliest design stage through development, content creation and production in a more connected workflow.
At the heart of that push is research such as GarmageNet, 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, the research tackles one of the trickier problems in fashion AI: figuring out how a flat 2D pattern translates into an actual three-dimensional garment.
GarmageNet brings pattern outlines, 3D geometry and sewing information together in a single representation.
The system can take a range of inputs - including text, sketches, images, patterns and point clouds - and use them to generate structured assets, reconstruct sewing relationships and initialise physical simulation.
In practical terms, that could allow designers to explore ideas, create digital samples and make edits without treating AI-generated imagery as the end of the process.
Instead, the technology links the original input to the underlying pattern, the way pieces are sewn together and the physical simulation of the garment - bringing AI closer to the wider product-development workflow.
The research is supported by GarmageSet, a dataset containing 14,801 professionally created garments.
And the numbers from the research are striking.
In the paper’s evaluations, GarmageJigsaw achieved 99.16% precision and 97.13% recall when identifying sewing points.
GarmageNet achieved a 91.41% simulation-initialisation success rate on 150 complex patterns, with a reported inference time of approximately eight seconds per garment under the study’s test conditions.
The technology reflects Style3D’s background in deformable-object simulation, garment deformation, digital fabrics and AI-assisted workflows.
Rather than treating AI as a standalone tool, the company is building towards a system where different technologies and capabilities can work together across the fashion pipeline.
Style3D maintains a substantial, dedicated R&D organisation and had published more than 30 papers at conferences including SIGGRAPH, SIGGRAPH Asia, CVPR, NeurIPS and 3DV.
As of 28 July 2026, Style3D held 106 granted patents worldwide, including 58 invention patents and 12 granted in Europe and the United States, with a further 155 invention patent applications pending.
The company also operates an overseas R&D centre in Germany.
Style3D’s AI and 3D technologies are already being used across a range of customer workflows.
German menswear brand OLYMP uses Style3D and Assyst to connect 2D patterns, 3D simulation, visualisation and digital showrooms.
According to the company, this allows styles to be developed in days rather than weeks, while reducing physical samples and improving communication between design, sales and production teams.
And Style3D believes the next big step could come from AI agents — systems capable of connecting several different skills rather than simply generating an image or completing one isolated task.
As Style3D CEO Eric Liu told Just Style, “The most important development is the agent. We use agents to link different skills.”
The company is now advancing a connected fashion workflow in which AI agents can coordinate product data, 3D design, simulation, content creation and production preparation — while human teams retain creative and operational responsibility.
For fashion, that could mean AI's role is no longer simply about dreaming up what a garment might look like.
It could increasingly be about helping take that idea from the first sketch to something ready to make.
From a flat pattern to a virtual garmentThe bigger AI fashion pictureWhat this looks like in the real world