GH-NAF：面向差异感知CBCT的网格自适应哈希层级注意力神经衰减场
GH-NAF: Grid-Adaptive Hash-Level-Attended Neural Attenuation Fields for Discrepancy-Aware CBCT
把多分辨率哈希编码的各层级解耦，并用不确定性引导的层级注意力自适应加权，这一思路可直接迁移到稀疏视角NeRF/3DGS重建中处理投影噪声与不一致性，缓解均匀组织中的伪高频纹理与边界模糊。
原文依据 decouples hash levels and adaptively weights them via uncertainty-guided, grid-adaptive hash-level attention
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论基于神经辐射场（NeRF）并结合多分辨率哈希编码的方法能够实现高效的稀疏视角CBCT重建，但真实投影因散射/噪声及相关不一致性而违背理想假设。均匀融合哈希网格各层级会纠缠异质频率成分，在同质组织中产生虚假高频纹理、边界模糊，并传播投影引入的偏差。我们提出GH-NAF（网格自适应哈希层级注意力神经衰减场），它解耦哈希层级，并通过不确定性引导的、网格自适应的哈希层级注意力对其进行自适应加权。这既稳定了同质区域中的低频建模，又在结构边界附近选择性地保留高频细节。在合成与真实CBCT数据上的实验表明，GH-NAF在同质材料内部对比度和重建质量上优于最优方法。代码见 https://github.com/seongje-oh/GH-NAF。
英文原文摘要 994 字符
Neural radiance fields (NeRF)-based methods with multi-resolution hash encoding enable efficient sparse-view CBCT reconstruction, but real-world projections violate ideal assumptions due to scatter/noise and related inconsistencies. Uniformly fusing hash-grid levels entangles heterogeneous frequency components, yielding spurious high-frequency textures in homogeneous tissues, blurred boundaries, and propagation of projection-induced bias. We propose GH-NAF (Grid-Adaptive Hash-Level-Attended Neural Attenuation Field), which decouples hash levels and adaptively weights them via uncertainty-guided, grid-adaptive hash-level attention. This stabilizes low-frequency modeling in homogeneous regions while selectively preserving high-frequency details near structural boundaries. Experiments on synthetic and real CBCT data show that GH-NAF improves intra-material contrast and reconstruction quality over state-of-the-art methods. The code is available at https://github.com/seongje-oh/GH-NAF