无需标注平面，仅用两张图就能重建室内场景的精确3D平面结构。
PLANA3R: Zero-shot Metric Planar 3D Reconstruction via Feed-Forward Planar Splatting
- 用视觉变换器提取稀疏平面，通过平面点云渲染反向传播梯度。
- 在无平面标注下训练，在多个数据集上实现毫米级精度重建。
- 适合做室内3D重建、平面分割和相机位姿估计的开发者使用。
本文针对室内场景的度量3D重建问题，利用其固有的几何规律，提出基于平面3D基元的紧凑表示方法。我们引入PLANA3R，一种无需相机姿态的两视图度量平面3D重建框架。该方法利用视觉变换器提取稀疏平面基元，估计相对相机位姿，并通过平面点云投影生成高分辨率深度与法线图进行几何学习监督，梯度可沿渲染路径反向传播。不同于以往前馈方法需依赖3D平面标注训练，PLANA3R在无显式平面监督下学习平面3D结构，可在大规模立体图像数据集上仅使用深度与法线标注进行可扩展训练。我们在多个室内场景数据集上验证了PLANA3R的性能，在度量评估协议下展现出对域外环境的强泛化能力，涵盖3D表面重建、深度估计和相对位姿估计等任务。此外，基于平面3D表示，本方法还自然具备准确的平面分割能力。
原文摘要 · Abstract (English)
This paper addresses metric 3D reconstruction of indoor scenes by exploiting their inherent geometric regularities with compact representations. Using planar 3D primitives - a well-suited representation for man-made environments - we introduce PLANA3R, a pose-free framework for metric Planar 3D Reconstruction from unposed two-view images. Our approach employs Vision Transformers to extract a set of sparse planar primitives, estimate relative camera poses, and supervise geometry learning via planar splatting, where gradients are propagated through high-resolution rendered depth and normal maps of primitives. Unlike prior feedforward methods that require 3D plane annotations during training, PLANA3R learns planar 3D structures without explicit plane supervision, enabling scalable training on large-scale stereo datasets using only depth and normal annotations. We validate PLANA3R on multiple indoor-scene datasets with metric supervision and demonstrate strong generalization to out-of-domain indoor environments across diverse tasks under metric evaluation protocols, including 3D surface reconstruction, depth estimation, and relative pose estimation. Furthermore, by formulating with planar 3D representation, our method emerges with the ability for accurate plane segmentation. The project page is available at https://lck666666.github.io/plana3r
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。