A deep learning approach to numerically approximate the solution to the Eikonal equation is introduced. The proposed method is built on the fast marching scheme which comprises of two components: a local numerical solver and an update scheme. We replace the formulaic local numerical solver with a trained neural network to provide highly accurate estimates of local distances for a variety of different geometries and sampling conditions. Our learning approach generalizes not only to flat Euclidean domains but also to curved surfaces enabled by the incorporation of certain invariant features in the neural network architecture. We show a considerable gain in performance, validated by smaller errors and higher orders of accuracy for the numerical solutions of the Eikonal equation computed on different surfaces The proposed approach leverages the approximation power of neural networks to enhance the performance of numerical algorithms, thereby, connecting the somewhat disparate themes of numerical geometry and learning.
A fast eikonal equation solver using the Schrodinger wave equation
International Center for Theoretical Sciences, Tata Institute of Fundamental Research(塔塔基础研究院国际理论科学中心);Florida Institute of Technology(佛罗里达理工学院);University of Florida(佛罗里达大学)
DeepShapeMatchingKit: Accelerated Functional Map Solver and Shape Matching Pipelines Revisited
Technical University of Munich(慕尼黑工业大学);Munich Center for Machine Learning(慕尼黑机器学习中心);MIT(麻省理工学院);Adobe;Imperial College London(帝国理工学院);Simon Fraser University(西蒙菲莎大学)
University of Toronto(多伦多大学);Uber Advanced Technologies Group(优步先进技术集团);Massachusetts Institute of Technology(麻省理工学院);University of California San Diego(加利福尼亚大学圣迭戈分校)