PartDiffuser：基于离散扩散的部件级三维网格生成
PartDiffuser: Part-wise 3D Mesh Generation via Discrete Diffusion
把「部件间自回归 + 部件内并行离散扩散」的混合生成范式用于网格生成，并用点云作为分层几何条件通过部件感知交叉注意力注入，为可控、高细节的3D资产生成提供了可复用的条件控制与解耦思路。
原文依据 employs autoregression between parts to ensure global topology, while utilizing a parallel discrete diffusion process within each semantic part
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论现有的用于生成艺术家设计网格的自回归（AR）方法难以在全局结构一致性与高保真局部细节之间取得平衡，且容易受到误差累积的影响。为解决这一问题，我们提出 PartDiffuser，一种用于点云到网格生成的新型半自回归扩散框架。该方法首先对网格进行语义分割，然后以「部件级」方式运作：在部件之间采用自回归以保证全局拓扑，同时在每个语义部件内部利用并行的离散扩散过程来精确重建高频几何特征。PartDiffuser 基于 DiT 架构，并引入部件感知交叉注意力机制，使用点云作为分层几何条件来动态控制生成过程，从而有效解耦全局与局部的生成任务。实验表明，该方法在生成细节丰富的三维网格方面显著优于最优（SOTA）模型，展现出适用于真实世界应用的卓越细节表现能力。
英文原文摘要 1136 字符
Existing autoregressive (AR) methods for generating artist-designed meshes struggle to balance global structural consistency with high-fidelity local details, and are susceptible to error accumulation. To address this, we propose PartDiffuser, a novel semi-autoregressive diffusion framework for point-cloud-to-mesh generation. The method first performs semantic segmentation on the mesh and then operates in a "part-wise" manner: it employs autoregression between parts to ensure global topology, while utilizing a parallel discrete diffusion process within each semantic part to precisely reconstruct high-frequency geometric features. PartDiffuser is based on the DiT architecture and introduces a part-aware cross-attention mechanism, using point clouds as hierarchical geometric conditioning to dynamically control the generation process, thereby effectively decoupling the global and local generation tasks. Experiments demonstrate that this method significantly outperforms state-of-the-art (SOTA) models in generating 3D meshes with rich detail, exhibiting exceptional detail representation suitable for real-world applications.