用少于100参数实现可编辑的轻量级3D形状表示
Fluid-SDF: Ultra-Lightweight and Editable Implicit Shape Representation via Differentiable Primitives
- 用可微分几何体混合替代传统神经网络，实现紧凑形状建模
- 仅用不到100个参数就达到与神经基线相当的形状重建精度
- 支持零样本用户直接编辑形状，适合移动端和嵌入式设备
隐式神经表示（INRs）已成为连续2D形状建模的标准，但存在黑箱难编辑、对噪声敏感及参数量大等问题，严重限制其在边缘设备上的部署。我们提出Fluid-SDF，一种高度压缩的可微分构造实体几何（CSG）框架，通过平滑最小函数混合显式几何体来建模形状。通过用参数化几何体引擎替代传统多层感知机（MLPs），Fluid-SDF以严格少于100个参数重建复杂非凸拓扑，达到或优于标准神经基线的平均交并比（mIoU）。此外，我们证明Fluid-SDF作为强大几何先验，能有效抵抗高频数据噪声，而容量匹配的神经网络则会灾难性过拟合。不同于标准INRs，Fluid-SDF的显式参数空间支持无需重训练的零样本用户编辑，可直接修改局部与全局形状特征。完全避免设备端昂贵的梯度更新，使其特别适用于移动AI、增强现实及资源受限的嵌入式环境。
原文摘要 · Abstract (English)
Implicit Neural Representations (INRs) have become the standard for continuous 2D shape modeling, but they suffer from black-box uneditability, vulnerability to noise, and high parameter counts that severely hinder deployment on edge devices. We introduce Fluid-SDF, a highly compressed, differentiable Constructive Solid Geometry (CSG) framework that models shapes using explicit geometric primitives blended via a smooth minimum function. By replacing traditional multi-layer perceptrons (MLPs) with a parameterized primitive engine, Fluid-SDF reconstructs complex, non-convex topologies using strictly under 100 parameters, achieving comparable or superior intersection-over-union (mIoU) to standard neural baselines. Furthermore, we demonstrate that Fluid-SDF acts as a powerful geometric prior, inherently resisting high-frequency dataset noise where capacity-matched neural networks catastrophically overfit. Finally, unlike standard INRs, Fluid-SDF's explicit parameter space allows for direct, zero-shot user editing of local and global shape features without retraining. By bypassing expensive on-device gradient updates entirely, Fluid-SDF is uniquely suited for mobile AI, augmented reality, and resource-constrained embedded environments
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。