Deep Accurate Solver for the Geodesic Problem

TL;DR

This paper introduces a deep learning-based geodesic distance solver that achieves third-order accuracy on surfaces, surpassing traditional methods and previous learning approaches in precision.

Contribution

The paper presents a novel neural network-based local solver for geodesic distances that improves accuracy to third-order and offers a bootstrapping method for further enhancement.

Findings

The deep learning method outperforms classical polyhedral approximations.

The proposed solver achieves third-order accuracy.

Numerical experiments demonstrate superior precision over prior methods.

Abstract

A common approach to compute distances on continuous surfaces is by considering a discretized polygonal mesh approximating the surface and estimating distances on the polygon. We show that exact geodesic distances restricted to the polygon are at most second-order accurate with respect to the distances on the corresponding continuous surface. By order of accuracy we refer to the convergence rate as a function of the average distance between sampled points. Next, a higher-order accurate deep learning method for computing geodesic distances on surfaces is introduced. Traditionally, one considers two main components when computing distances on surfaces: a numerical solver that locally approximates the distance function, and an efficient causal ordering scheme by which surface points are updated. Classical minimal path methods often exploit a dynamic programming principle with quasi-linear computational complexity in the number of sampled points. The quality of the distance approximation is determined by the local solver that is revisited in this paper. To improve state of the art accuracy, we consider a neural network-based local solver which implicitly approximates the structure of the continuous surface. We supply numerical evidence that the proposed learned update scheme provides better accuracy compared to the best possible polyhedral approximations and previous learning-based methods. The result is a third-order accurate solver with a bootstrapping-recipe for further improvement.