按需路由：DSNet 实现高效渐进式点云去噪
Routing on Demand: DSNet for Efficient Progressive Point Cloud Denoising
把「按需路由」引入点云渐进去噪：用基于法向相似性的噪声判别器为每个图像块（patch）选择去噪模块，并用路径选择性迭代实现跨阶段跳连，可直接迁移到其他迭代式点云/几何处理流程中以减少冗余计算并缓解过平滑。
原文依据 enables routing on demand by adaptively determining the optimal denoising path for each local patch based on its noise characteristics
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论点云去噪是提升三维感知系统可靠性与准确性的关键预处理步骤。现有大多数渐进式去噪方法依赖固定的迭代流程，对所有区域一视同仁地处理，导致在处理噪声分布不均匀的点云时出现冗余计算，并使几何细节过度平滑。为克服这些局限，本文提出动态跳连网络（Dynamic Skip Net，DSNet），一种新颖的渐进式去噪框架，它依据每个局部图像块的噪声特性自适应地确定最优去噪路径，从而实现按需路由。DSNet 包含一个噪声判别器，通过分析法向相似性来量化局部噪声强度，并包含一个反向单调决策函数，将该度量映射到合适的去噪模块。此外，本文提出路径选择性迭代机制，在每个阶段动态重新评估恢复状态并重新规划去噪路线，从而实现跨阶段跳连，以尽量减少不必要的计算。在多个基准上的大量实验表明，DSNet 在噪声抑制、几何保真度和计算效率方面均达到最优性能。代码已发布于 https://github.com/cz-61/DSNet。
英文原文摘要 1302 字符
Point cloud denoising is a critical preprocessing step for enhancing the reliability and accuracy of 3D perception systems. Most existing progressive denoising methods rely on fixed iterative pipelines that process all regions uniformly, resulting in redundant computation and over-smoothing of geometric details when handling point clouds with non-uniform noise distributions. To overcome these limitations, we introduce Dynamic Skip Net (DSNet), a novel progressive denoising framework that enables routing on demand by adaptively determining the optimal denoising path for each local patch based on its noise characteristics. DSNet incorporates a noise discriminator that quantifies local noise intensity by analyzing normal similarity, and a reverse monotonic decision function that maps this measure to an appropriate denoising module. Furthermore, we propose a path-selective iteration mechanism that dynamically re-evaluates the restoration state and re-plans the denoising route at each stage, enabling cross-stage skipping to minimize unnecessary computation. Extensive experiments on multiple benchmarks demonstrate that DSNet achieves state-of-the-art performance in noise suppression, geometric fidelity, and computational efficiency. The code is available at https://github.com/cz-61/DSNet