Subdguide improves mesh to subdivision surface reconstruction accuracy and repair
SubDGuide: A Modeler-Inspired Agentic Workflow for Mesh-to-SubD Reconstruction
Graphics
Summary
Subdivision surfaces simplify complex shapes by using a small set of control points that define a smooth surface. Turning a detailed mesh back into this smaller control structure is tricky because the system needs to decide which parts of the shape need control points and how to fix errors during the process. The authors created SubDGuide, a step-by-step method inspired by how human modelers work, which plans the reconstruction, checks the results, and makes corrections as needed. Their method produces more accurate and higher-quality results than previous automatic methods.
What this means in practice
- •For 3d modelers: Improve mesh-to-subdivision workflows by automatically planning and repairing control cages for complex shapes using geometry and semantic feedback.
- •For computer graphics software developers: Integrate an inspectable and interactive mesh-to-SubD reconstruction pipeline that supports multimodal planners and verification to enhance modeling tools.$Commercial implications: Enables development of advanced modeling software with automated reconstruction and repair features that can be sold to 3D content creators.
Authors
Mengnan Jiang, Christian Franke, Michele Franco Adesso, Antonio Haas, Grace Li Zhang
Abstract
Subdivision surfaces represent free-form geometry through a sparse control cage, but recovering that cage from a dense mesh is not merely a fitting problem: the system must infer where control is needed, which curves encode design features, and when an initial result should be revised. We present SubDGuide, a modeler-inspired agentic workflow for this task. Stage A reads aligned multiview geometric evidence and produces a compact plan for cage resolution, feature mapping, and broad-form fitting. Stage B inspects the resulting surface, requests targeted diagnostics, and chooses repair, rollback, or stopping actions. Geometry tools execute and verify every change; the planner never generates vertices or connectivity. On a fixed evaluation cohort, SubDGuide improves all five reported metrics over automatic remeshing: median Chamfer-L1 decreases from 0.641% to 0.443% of the target bounding-box diagonal, and F-score at a 1% tolerance increases from 82.00% to 92.78%. Stateful feedback improves four of five reported median metrics over one-shot planning, while verification retains the earlier checkpoint when a proposal is unhelpful. The same interface supports six multimodal planners, showing how semantic judgment can guide a practical, inspectable mesh-to-SubD workflow.