iLRM：迭代式大型三维重建模型
iLRM: An Iterative Large 3D Reconstruction Model
把多视图全局注意力拆成两阶段注意力、并把场景表示与输入图像解耦，是让前馈 3DGS 重建在视图数与分辨率增大时仍可扩展的通用手段，可直接迁移到任何多视图 Transformer 重建或新视角合成管线。
原文依据 decomposing global multi-view interactions into a two-stage attention scheme to reduce computational costs
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论前馈三维建模已成为快速、高质量三维重建的一种有前景的途径。特别是直接生成显式三维表示（如三维高斯泼溅）的方法，因其渲染快速且质量高而受到广泛关注。然而，许多最优方法主要基于 Transformer 架构，由于它们依赖对来自多个输入视图的图像 token 做全注意力，随着视图数量或图像分辨率的增加，计算开销变得难以承受，因而存在严重的可扩展性问题。为实现可扩展且高效的前馈三维重建，我们提出迭代式大型三维重建模型（iLRM），通过迭代细化机制生成三维高斯表示，并遵循三条核心原则：（1）将场景表示与输入图像解耦，以获得紧凑的三维表示；（2）将全局多视图交互分解为两阶段注意力方案，以降低计算开销；（3）在每一层注入高分辨率信息，以实现高保真重建。在 RE10K 和 DL3DV 等广泛使用的数据集上的实验结果表明，iLRM 在重建质量与速度上均优于现有方法。
英文原文摘要 1307 字符
Feed-forward 3D modeling has emerged as a promising approach for rapid and high-quality 3D reconstruction. In particular, directly generating explicit 3D representations, such as 3D Gaussian splatting, has attracted significant attention due to its fast and high-quality rendering. However, many state-of-the-art methods, primarily based on transformer architectures, suffer from severe scalability issues because they rely on full attention across image tokens from multiple input views, resulting in prohibitive computational costs as the number of views or image resolution increases. Toward a scalable and efficient feed-forward 3D reconstruction, we introduce an iterative Large 3D Reconstruction Model (iLRM) that generates 3D Gaussian representations through an iterative refinement mechanism, guided by three core principles: (1) decoupling the scene representation from input images to enable compact 3D representations; (2) decomposing global multi-view interactions into a two-stage attention scheme to reduce computational costs; and (3) injecting high-resolution information at every layer to achieve high-fidelity reconstruction. Experimental results on widely used datasets, such as RE10K and DL3DV, demonstrate that iLRM outperforms existing methods in both reconstruction quality and speed.