Neural Splines: Fitting 3D Surfaces with Infinitely-Wide Neural Networks

2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021(2021)

引用 45|浏览162
暂无评分
摘要
We present Neural Splines, a technique for 3D surface reconstruction that is based on random feature kernels arising from infinitely-wide shallow ReLU networks. Our method achieves state-of-the-art results, outperforming recent neural network-based techniques and widely used Poisson Surface Reconstruction (which, as we demonstrate, can also be viewed as a type of kernel method). Because our approach is based on a simple kernel formulation, it is easy to analyze and can be accelerated by general techniques designed for kernel-based learning. We provide explicit analytical expressions for our kernel and argue that our formulation can be seen as a generalization of cubic spline interpolation to higher dimensions. In particular, the RKHS norm associated with Neural Splines biases toward smooth interpolants.
更多
查看译文
关键词
fitting 3D surfaces,infinitely-wide neural networks,3D surface reconstruction,random feature kernels,infinitely-wide shallow ReLU networks,recent neural network-based techniques,Poisson surface reconstruction,kernel method,simple kernel formulation,general techniques,cubic spline interpolation,neural splines biases
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要