MeshRipple: Structured Autoregressive Generation of Artist-Meshes

TL;DR

MeshRipple is a novel method for 3D mesh generation that expands meshes from a surface frontier, effectively capturing long-range dependencies and producing high-quality, topologically complete meshes.

Contribution

It introduces a frontier-aware BFS tokenization, an expansive prediction strategy, and a sparse-attention global memory to improve autoregressive mesh generation.

Findings

Outperforms recent baselines in mesh quality.

Produces meshes with high surface fidelity.

Achieves topological completeness in generated meshes.

Abstract

Meshes serve as a primary representation for 3D assets. Autoregressive mesh generators serialize faces into sequences and train on truncated segments with sliding-window inference to cope with memory limits. However, this mismatch breaks long-range geometric dependencies, producing holes and fragmented components. To address this critical limitation, we introduce MeshRipple, which expands a mesh outward from an active generation frontier, akin to a ripple on a surface. MeshRipple rests on three key innovations: a frontier-aware BFS tokenization that aligns the generation order with surface topology; an expansive prediction strategy that maintains coherent, connected surface growth; and a sparse-attention global memory that provides an effectively unbounded receptive field to resolve long-range topological dependencies. This integrated design enables MeshRipple to generate meshes with high surface fidelity and topological completeness, outperforming strong recent baselines.