Opti-NeuS：面向双层透明与不透明物体的神经重建
Opti-NeuS: Neural Reconstruction for Dual-Layered Transparent and Opaque Objects
用网络预测空间变化的折射率来追踪折射光线路径，从而把透明物体的折射畸变纳入神经隐式重建；两阶段分层解耦加 alpha 混合分离内外表面，可直接迁移到透明容器、玻璃、液体等混合材质的重建与渲染。
原文依据 Opti-NeuS incorporates a novel Network to obtain spatially-varying IoR for tracing the refractive ray paths
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论从多视角对透明物体进行三维重建一直是一项长期挑战。与不透明物体不同，透明物体会产生复杂的折射，导致严重的图像畸变，从而使问题高度病态。现有重建方法通常依赖特殊采集设备或受控环境，以提供更多先验并简化折射建模。更重要的是，这些方法缺乏对透明与不透明混合物体的重建能力，仅局限于透明或不透明材质。为应对这些挑战，我们提出 Opti-NeuS，一种无需受控环境或额外输入即可重建透明与不透明物体的新方法。Opti-NeuS 引入一个新的网络来获得空间变化的折射率，用于追踪折射光线路径，最终可对折射造成的视觉畸变进行建模。为处理双层透明与不透明物体，我们设计了一种两阶段分层重建策略，将外层与内层几何解耦，并结合 alpha 混合实现感知透明性的表面分离。实验表明，Opti-NeuS 具有实际有效性，并优于先前工作。
英文原文摘要 1244 字符
3D reconstruction of transparent objects from multiple views has been a long-standing challenge. In contrast to opaque objects, transparent objects exhibit complex refraction that causes serious image distortions, resulting in a highly ill-posed problem. Existing reconstruction methods commonly depend on special capture devices or controlled environments, which provide more priors and simplify the modeling of refraction. More importantly, these methods lack the capability for reconstruction of mixed transparent and opaque objects, being confined to transparent or opaque materials. To address these challenges, we propose Opti-NeuS, a novel method for reconstructing transparent and opaque objects without controlled environments or additional input. Opti-NeuS incorporates a novel Network to obtain spatially-varying IoR for tracing the refractive ray paths, which can finally model refractive visual distortions. To deal with dual-layered transparent and opaque objects, we devise a two-stage hierarchical reconstruction strategy that decouples outer and inner geometry, combined with alpha-blending for transparency-aware surface separation. Experiments show that Opti-NeuS achieves practical effectiveness and outperforms prior works.