DressAgent reconstructs layered 3D garments from single images fast
Inverse Garment Design with Visual Agents
Artificial Intelligence
Summary
This paper focuses on creating 3D models of clothes with multiple layers, all from just one picture. The authors propose DressAgent, a system that plans and improves garment patterns using virtual sewing simulations instead of slow optimization or limited training data. They built a large dataset of sewing patterns to help train and test their system. DressAgent works step-by-step, first guessing coarse layers and then refining each pattern with a set of virtual 'agents' that simulate and judge the results to match the picture closely. It works faster and better than previous methods, handling complex clothing layers and different body shapes.
What this means in practice
- •For video game developers: Create realistic 3D outfits from concept images quickly to enhance character customization with detailed layered clothing.$Commercial implications: Enables game studios to generate complex virtual costumes for characters efficiently, reducing manual modeling labor and accelerating production.
- •For fashion designers: Convert single photos into sewing patterns and 3D garment simulations to prototype layered outfits virtually before physical sampling.
Authors
Nanxiang Jiang, Yunuo Chen, Zeng Tao, Zizhang Li, Qi Guo, Jiajun Wu, Yin Yang, Chenfanfu Jiang
Abstract
We study the problem of inverse garment design, reconstructing simulation-ready 3D outfits with arbitrary layers from a single image, with visual agents. Unlike differentiable optimization that is time-consuming, or feed-forward methods limited by training data, we introduce DressAgent, a hierarchical agentic system that recovers garment sewing patterns through planning and closed-loop refinement grounded in physical simulation. To this end, we first present an enhanced representation based on GarmentCode for diverse styles and arbitrary dressing layers, and build the first multi-layer sewing pattern dataset with over 15K outfits for supervision and benchmarking. Inverse design then proceeds coarse to fine: In layer planning and coarse estimation, agents infer layer order and relations, peel each layer into its own view, and estimate coarse patterns. In fine-grained refinement, an agentic trio of Tailor, Critic, and Orchestrator refines patterns in a closed loop of editing, simulating, and judging against the reference with visual and measured evidence. Carefully designed tools and an evolving memory make this refinement converge on the target geometry. We also design an agentic generation method for high-fidelity textures. DressAgent outperforms feed-forward methods by a large margin, surpasses or matches differentiable optimization in a fraction of the time, and generalizes to complex layering, diverse poses, and body shapes.