用可渲染场指导高斯点云，提升复杂场景的视角生成稳定性。
Rendering Anywhere You See: Renderability Field-guided Gaussian Splatting
- 通过可渲染场量化输入不均匀性，指导伪视角采样
- 训练图像修复模型提升远基线伪视角的视觉质量
- 适合需要稳定多视角生成的虚拟/增强现实应用
场景视角合成在虚拟现实、增强现实和机器人等领域日益重要。与仅针对物体的任务（如汽车360°视图生成）不同，场景视角合成需处理整个环境，非均匀观测带来渲染质量不稳定的问题。为此，我们提出一种新方法：可渲染场引导的高斯点云渲染（RF-GS）。该方法通过可渲染场量化输入的不均匀性，指导伪视角采样以提升视觉一致性。为保证宽基线伪视角的质量，我们训练了一个图像修复模型，将点投影映射到可见光风格。此外，经验证的混合数据优化策略有效融合了伪视角角度与源视角纹理信息。在模拟和真实数据上的对比实验表明，本方法在渲染稳定性方面优于现有方法。
原文摘要 · Abstract (English)
Scene view synthesis, which generates novel views from limited perspectives, is increasingly vital for applications like virtual reality, augmented reality, and robotics. Unlike object-based tasks, such as generating 360° views of a car, scene view synthesis handles entire environments where non-uniform observations pose unique challenges for stable rendering quality. To address this issue, we propose a novel approach: renderability field-guided gaussian splatting (RF-GS). This method quantifies input inhomogeneity through a renderability field, guiding pseudo-view sampling to enhanced visual consistency. To ensure the quality of wide-baseline pseudo-views, we train an image restoration model to map point projections to visible-light styles. Additionally, our validated hybrid data optimization strategy effectively fuses information of pseudo-view angles and source view textures. Comparative experiments on simulated and real-world data show that our method outperforms existing approaches in rendering stability.
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。