SASNet：用于隐式神经表示的空间自适应正弦网络
SASNet: Spatially-Adaptive Sinusoidal Networks for INRs
SASNet 通过冻结频率嵌入层和可学习空间掩码解决 SIREN 训练不稳定和频率初始化敏感问题，可直接用于图像/表面重建、SDF 重建等 INR 任务，提升收敛速度和重建质量。
原文依据 we propose SASNet, a Spatially-Adaptive Sinusoidal Network that couples a frozen frequency embedding layer
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论正弦神经网络（SIREN）是视觉和图形学中用于低维信号的强大隐式神经表示（INR）。通过用正弦函数编码输入坐标，它们能够实现高频图像和表面重建。然而，训练 SIREN 往往不稳定且对频率初始化高度敏感：小频率会在细节区域产生过于平滑的重建，而大频率会引入虚假的高频成分，在图像背景等平滑区域表现为噪声。为了解决这些挑战，我们提出了 SASNet，一种空间自适应正弦网络，它将显式固定网络频率支撑的冻结频率嵌入层与联合学习的空间掩码相结合，后者在域内定位神经元的影响。这种配对稳定了优化，锐化了边缘，并抑制了平滑区域的噪声。在 2D 图像和 3D 体数据拟合以及符号距离场（SDF）重建基准上的实验表明，SASNet 实现了更快的收敛、更优的重建质量和鲁棒的频率定位——将低频和高频神经元分别分配给平滑和细节区域——同时保持参数效率。
英文原文摘要 1295 字符
Sinusoidal neural networks (SIRENs) are powerful implicit neural representations (INRs) for low-dimensional signals in vision and graphics. By encoding input coordinates with sinusoidal functions, they enable high-frequency image and surface reconstruction. However, training SIRENs is often unstable and highly sensitive to frequency initialization: small frequencies produce overly smooth reconstructions in detailed regions, whereas large ones introduce spurious high-frequency components that manifest as noise in smooth areas such as image backgrounds. To address these challenges, we propose SASNet, a Spatially-Adaptive Sinusoidal Network that couples a frozen frequency embedding layer, which explicitly fixes the network's frequency support, with jointly learned spatial masks that localize neuron influence across the domain. This pairing stabilizes optimization, sharpens edges, and suppresses noise in smooth areas. Experiments on 2D image and 3D volumetric data fitting as well as signed distance field (SDF) reconstruction benchmarks demonstrate that SASNet achieves faster convergence, superior reconstruction quality, and robust frequency localization--assigning low- and high-frequency neurons to smooth and detailed regions respectively--while maintaining parameter efficiency.