用全景补全增强稀疏视角3D生成，让画面更真实。
PanoPlane: Plane-Aware Panoramic Completion for Sparse-View Indoor 3D Gaussian Splatting
- 通过全景补全提供完整空间布局，指导生成过程。
- 仅需3张输入图，合成效果比现有方法高17.8% PSNR。
- 无需训练，推理时动态引导注意力，避免胡编乱造。
我们提出PanoPlane，一种面向稀疏视角室内新视角合成的高保真方法，通过全景场景补全重建闭合房间几何结构。与基于透视的方法不同，PanoPlane利用$360^{ ext{°}}$全景补全，在生成过程中以完整空间布局为条件。我们提出无训练的布局锚定注意力引导机制（Layout Anchored Attention Steering），在推理时将扩散模型内部表示中的注意力导向检测到的平面表面。通过将每个未观测区域的注意力引导至几何一致的已观测内容，该方法将无约束幻觉替换为有依据的表面外推。生成的全景补全为3D Gaussian Splatting提供监督，使仅凭三张输入图即可实现未观测区域的准确新视角合成。在Replica、ScanNet++和Matterport3D上的实验表明，该方法在3、6、9张输入图下均达到当前最佳的新视角合成质量，相较现有最优基线在PSNR上最高提升17.8%，且无需对扩散模型进行任何训练或微调。
原文摘要 · Abstract (English)
We present PanoPlane, an approach for high-fidelity sparse-view indoor novel view synthesis that reconstructs closed room geometry via panoramic scene completion. Unlike perspective-based methods that generate training views from limited fields of view, PanoPlane leverages $360^{\circ}$ panoramic completion to condition the generative process on the full spatial layout. We propose Layout Anchored Attention Steering, a training-free mechanism that steers attention within the diffusion model's internal representation toward scene's detected planar surfaces at inference time. By directing each unobserved region's attention toward geometrically consistent observed content, our method replaces unconstrained hallucination with grounded surface extrapolation. The resulting panoramic completions provide supervision for 3D Gaussian Splatting, enabling accurate novel-view synthesis across unobserved regions from as few as three input views. Experiments on Replica, ScanNet++, and Matterport3D demonstrate state-of-the-art novel view synthesis quality across 3, 6, and 9 input views, achieving up to $+17.8\%$ improvement in PSNR over the current state-of-the-art baseline without any training or fine-tuning of the diffusion model.
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。