🤖 AI Summary
Existing methods for 3D mesh seam generation rely on projection or heuristic alignment, often producing artifacts and lacking robustness. This work proposes the first mesh-native autoregressive framework for seam generation, which directly traces seam paths vertex-by-vertex on the mesh graph, achieving edge alignment without any projection. The core innovations include a hierarchical sequential representation termed ChainingSeams, a dual-stream encoder that fuses topological and geometric information, and a Transformer-based autoregressive pointer layer that enables fine-grained seam planning within local neighborhoods under global structural guidance. Experimental results demonstrate that the proposed method significantly outperforms existing optimization- and learning-based baselines in terms of seam coherence, fidelity, and robustness.
📝 Abstract
We present MeshTailor, the first mesh-native generative framework for synthesizing edge-aligned seams on 3D surfaces. Unlike prior optimization-based or extrinsic learning-based methods, MeshTailor operates directly on the mesh graph, eliminating projection artifacts and fragile snapping heuristics. We introduce ChainingSeams, a hierarchical serialization of the seam graph that prioritizes global structural cuts before local details in a coarse-to-fine manner, and a dual-stream encoder that fuses topological and geometric context. Leveraging this hierarchical representation and enriched vertex embeddings, our MeshTailor Transformer utilizes an autoregressive pointer layer to trace seams vertex-by-vertex within local neighborhoods, ensuring projection-free, edge-aligned seams. Extensive evaluations show that MeshTailor produces more coherent, professional-quality seam layouts compared to recent optimization-based and learning-based baselines.