MatMart：基于扩散模型的 3D 物体材质重建
MatMart: Material Reconstruction of 3D Objects via Diffusion
把扩散模型用于 3D 物体的 PBR 材质重建，其视图-材质交叉注意力（VMCA）与渐进式推理让模型能接受任意数量输入视图，并借助生成先验补全未观测视角的材质，可直接迁移到逆渲染、重光照与资产重建流程。
原文依据 by utilizing progressive inference alongside the proposed view-material cross-attention (VMCA), MatMart enables reconstruction from an arbitrary number of input images
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论将扩散模型应用于基于物理的材质估计与生成近来备受关注。本文提出 MatMart，一种面向 3D 物体的新型材质重建框架，具有以下优势。首先，MatMart 采用两阶段重建：先从输入中准确预测材质，随后针对未观测视角进行先验引导的材质生成，从而获得高保真结果。其次，通过结合渐进式推理与所提出的视图-材质交叉注意力（VMCA），MatMart 能够从任意数量的输入图像进行重建，展现出很强的可扩展性与灵活性。最后，MatMart 通过对单个扩散模型进行端到端优化，同时实现材质预测与生成能力，无需依赖额外的预训练模型，因而在各类物体上表现出更强的稳定性。大量实验表明，与现有方法相比，MatMart 在材质重建上取得了更优的性能。
英文原文摘要 1070 字符
Applying diffusion models to physically-based material estimation and generation has recently gained prominence. In this paper, we propose MatMart, a novel material reconstruction framework for 3D objects, offering the following advantages. First, MatMart adopts a two-stage reconstruction, starting with accurate material prediction from inputs and followed by prior-guided material generation for unobserved views, yielding high-fidelity results. Second, by utilizing progressive inference alongside the proposed view-material cross-attention (VMCA), MatMart enables reconstruction from an arbitrary number of input images, demonstrating strong scalability and flexibility. Finally, MatMart achieves both material prediction and generation capabilities through end-to-end optimization of a single diffusion model, without relying on additional pre-trained models, thereby exhibiting enhanced stability across various types of objects. Extensive experiments demonstrate that MatMart achieves superior performance in material reconstruction compared to existing methods.